JP2018137241A - Technique for aligning-printing permanent layer with improved speed and accuracy - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a technique for aligning-printing a permanent layer with improved speed and accuracy.SOLUTION: In a reciprocative manufacturing process, a liquid for each product held by a substrate is accumulated, and a printer for forming a thin film is used. The liquid is dried so that a permanent layer of each product is formed by the liquid, is hardened, and or is handled differently. For printing, the substrate newly introduced is roughly and mechanically aligned with an optical system for detecting an error alignment less than millimeter and a software for correcting the error alignment. For providing a precise accumulation of the liquid for a predetermined region of the substrate, data rendering set so that a nozzle is assigned so as to accumulate an ink droplet regularly while depending on the error alignment.SELECTED DRAWING: Figure 2B

Description

(Cross-reference to related applications)
This application is filed on October 2, 2014 for the first inventor Michael Baker, US Provisional Application No. 121 and 59 for “Techniques For Arrayed Printing Of A Permanent Layered Improved Speed And Accuracy”, "Techniques For Arrayed Printing Of A Permanent" filed on July 7, 2014 for first inventor Michael Baker
Priority is claimed under each of US Provisional Application No. 62/021584 relating to “Layer With Impended Speed And Accuracy”. In addition, this application is also entitled “Techniques for Print Ink Volume Control Deposit Fluids Whistle Precision” filed on April 7, 2015 for the first inventor Naharid Harjee.
US utility patent application No. 14/680960 relating to Tolerances, filed for first inventor Naharid Harjee on July 3, 2014, entitled “Techniques for Print Ink Droplet measurement and Control to Deposition to Deposition to Deposition to Deposition. US Practical Application No. 14/340403 and “Encapsulation of Components of Electronic Device Using Half Control Tonic US Patent Application No. 14/340403, filed on Feb. 20, 2015 for first inventor Eliyahu Vronsky” 14/6 Claims priority based on 27186, and is a continuation-in-part of each of them.

U.S. Utility Patent Application No. 14/680960 is a continuation of U.S. Utility Patent Application No. 14/162525 (now U.S. Pat. No. 9,010,899 issued Apr. 21, 2015). In turn, U.S. Utility Patent Application No. 14/162525 was filed on December 26, 2013 by "Techniques for Print" by the first inventor, Najid Harjee.
Ink Volume Control To Deposit Fluids Within Precision Tolerances "Taiwan Patent Application No. 102148330 Claiming priority based on PCT patent application No. PCT / US2013 / 077720 filed for. PCT Patent Application No. PCT / US2013 / 077720 is a US Provisional Patent Application No. 61 / 746,545 for “Smart Mixing” filed on Dec. 27, 2012 for the first inventor Conor Francis Madigan; US Provisional Patent Application No. 61/822855 relating to “Systems and Methods Providing Uniform Printing of OLED Panels” filed on May 13, 2013 for the first inventor Nahid Harjee; first invention on July 2, 2013 United States provisional patent application relating to “Systems and Methods Providing Uniform Printing of OLED Panels” filed for the author Nahid Harjee 1/842351 issue; the first aspect of the present invention have Nahid on July 23, 2013
US Provisional Patent Application No. 61/857298 filed for “Systems and Methods Providing Uniform Printing of OLED Panels” filed for Harjee; “Systems” filed for first inventor Naked Harjee on November 1, 2013 and Methods Providing Uniform Printing of OLED Panels "and" Techniques for Print Volume Ink Volt Ink Volt Ink Volt Ink Volt Int Volt Ink Volt Int. " With Within Precision Tolerances " That claims priority to each of U.S. Provisional Patent Application No. 61/920715.

U.S. Utility Patent Application No. 14/340403, in turn, was filed on April 23, 2014 for the first inventor Naharid Harjee, “Techniques for Print Ink Drip Measurement and Control to Deposition Fluid PCs”. Patent application No. PCT / US2014 / 035193 and “Techniques for Print Ink Volume Control Deposit Fluids United Principal Patent Application No. 14 for the first inventor Naked Harjee” on Jan. 23, 2014 / 16525 Partial continuation applications for each Yes, and also filed for the first inventor Alexander Sou-Kang Ko on April 26, 2013, “OLED Printing Systems and Methods Using Laser Light Scattering for Measuring Drop Draft Vs. Patent application No. 61/816696, filed on August 14, 2013 for the first inventor Alexander Sou-Kang Ko, "OLED Printing Systems and Methods Usable Laser Light Scattering For Measuring In Measure "Techniques for Print Ink Volume Control To Deposit Fluids With Precision" by US Patent No. 61/866031 on December 26, 2013 and first inventor Naharid Harjee on Dec. 26, 2013.
Claims a benefit based on each of Taiwan patent application No. 102148330 filed for “Tolerances”.

  PCT Patent Application No. PCT / US2014 / 035193 was filed in order for the first inventor Alexander Sou-Kang Ko on April 26, 2013, "OLED Printing Systems and Methods Uster Laser Dancing Scattering Maker". United States Provisional Patent Application No. 61/816696 relating to “Size, Velocity and Trajectory”, “OLED Printing Systems and Methods Using Methods Us Method” filed for the first inventor Alexander Sou-Kang Ko on August 14, 2013. Ink US Provisional Patent Application No. 61/866031 on “Drop Size, Velocity and Trajectory”, and “Techniques for Print InPink Ink Volume Control”, filed on January 23, 2014 for the first inventor Nahad Harjee. Claims priority based on US utility patent application No. 14/162525 for Tolerances.

  U.S. Utility Patent Application No. 14/627186 is a continuation of U.S. Utility Application No. 14 / 485,005 (now U.S. Pat. No. 8,995,522 issued March 31, 2015). U.S. Utility Patent Application No. 14 / 485,005, in turn, is a U.S. Provisional Patent Application No. 61 for "Ink-Based Layer Fabrication Using Halftoning Variation" filed on December 12, 2013 for the first inventor Eliyahu Vronski. No./915149, US Provisional Patent Application No. 61/9793939, May 2014 related to “Ink-Based Layer Fabrication Using Controlling Thickness” filed on April 10, 2014 for the first inventor, Eliyahu Vronski. "Ink-Based Layer Fabrica" filed for first inventor Eliyahu Vronski on 30th US Provisional Patent Application No. 62/005044 relating to “Tion Using Halftoning To Control Thickness” and “Ink-Basic Layer Fabrication Usting HalfTonflation Usting HalfTonflation Usting Halftonficing Usting” filed for first inventor Eliyahu Vronski on June 30, 2014. Claiming priority based on US Provisional Patent Application No. 62/019076.

  Priority is claimed for each of the aforementioned applications, and each of the aforementioned patent applications is hereby incorporated by reference.

(background)
Some manufacturing techniques use a printing process to deposit a layer of material on a substrate as part of the assembly process. For example, multiple solar panel or organic light emitting diode (OLED) displays can be manufactured together on a common glass or other substrate, so that the multiple panels ultimately create their respective devices. Is cut from the common substrate. The printing process has a solvent that suspends or carries the material from which each permanent layer is formed, for example by curing the liquid to a permanent form, drying or otherwise “treating” ( For example, a liquid (similar to “ink”) is deposited. The liquid is deposited for each product in a carefully controlled manner such that each deposition layer closely aligns with the underlying layer on the substrate and the desired product location in place. Such alignment can result in high density patterns of microelectronic or optical structures where each layer of each structure has carefully controlled dimensions (including thickness) when high manufacturing accuracy is required. This is particularly important when the above process is used to fabricate.

  Returning to the example of an OLED display for illustrative purposes, each flat panel device manufactured in parallel on a common substrate is typically fabricated in a fluid well that holds light generating elements and electrodes. The pixel color component is a special color. Each deposition layer helps determine the proper operation of the individual pixels, and the more accurate each deposition process is (and the better it is aligned and controlled), the smaller pixels can be created. In that particular size dimension, the operation of the finished optical or electrical structure is more reliable. Variations in a given panel (including thickness variations) between pixels are undesirable because they can cause visible defects in the finished product. For example, a typical OLED display (eg, used as an HDTV screen) can involve millions of pixels in a compact space, with only a small amount of liquid being deposited through the printing process. If it has even variations, this can potentially be seen as a difference in brightness or color by the human eye. Thus, for such manufacturing applications, for example, at micron or finer resolution, precise printer control is required with a maximum variation in total fluid deposition per area of less than a half percent.

  In addition, it is desirable to maximize manufacturing throughput in order to produce products at acceptable consumer prices. If a given manufacturing device (including, for example, an industrial printer) takes a significant amount of time for each layer, this translates into slower manufacturing and increased consumer prices, such increased prices being Threatens feasibility.

This specification provides the following items, for example.
(Item 1)
A method of making a flat panel device,
Receiving a first substrate;
Using a printer to print a liquid on the first substrate, the liquid carrying a material capable of forming a permanent layer of the flat panel device; and forming the permanent layer And then processing the liquid to
(A) the method further comprises detecting a position of a reference on the first substrate and detecting an error depending on a difference between the position of the reference and an expected feature position. Including
(B) using the printer so as to receive stored data and generate adjusted data defining where the permanent layer for the flat panel device is to be printed; Rendering the stored data in accordance with the error; and printing the liquid droplets on the first substrate in accordance with the adjusted data;
(C) rendering the stored data further comprises:
In a manner that the target total volume of the liquid deposited in a particular region is limited to be within a predetermined error threshold for the desired volume of liquid for the particular region,
Allocating individual nozzles of the printer printhead to deposit respective droplets of the liquid in the particular area of the first substrate, depending on the error.
(Item 2)
The substrate represents a respective product fabricated in parallel on the first substrate, and the flat panel device is one of the respective products;
Using the printer for printing comprises using the printer to print as part of an integral print and to produce each permanent layer of the respective product on the first substrate. Including
2. The method of item 1, further comprising separating the flat panel device from other respective products on the first substrate following the step of using a printer for the printing.
(Item 3)
Each of the respective products is an organic light emitting diode display device, and using the printer has a droplet density selected depending on the desired thickness of the permanent layer for the respective product. 3. The method of item 2, comprising using to print the liquid droplets on the substrate.
(Item 4)
Item 3. The method according to item 2, embodied as a method of producing a display for a flat panel television.
(Item 5)
The criterion comprises a first visual feature, the stored data is a first stored data, the error is a first error, and the adjusted data is a first Wherein the first visual feature is associated with the flat panel device but not with at least another one of the respective products;
The method further includes optically detecting a position of a second visible feature on the first substrate, and a first between the position of the second visible feature and an expected feature position. Detecting the second error, wherein the second visual feature is associated with the other one of the respective products, not the flat panel device,
Using the printer includes receiving second stored data defining where the permanent layer corresponding to the other one of the respective products is fabricated; Rendering the second stored data in accordance with the second error to produce adjusted data, the permanent for each of the other one of the flat panel device and the respective product. The first adjusted data according to the second adjusted data such that a liquid for a layer is printed on the first substrate in a manner dependent on the first error and the second error, respectively. 3. A method according to item 2, comprising printing on a substrate.
(Item 6)
Using the printer includes printing according to a scan path that represents a substantially continuous relative motion between the print head and the first substrate, the scan path comprising the flat panel device. The first error and the second, depending on whether they are currently traversing a location associated with each permanent layer for each of the permanent layers or the other one of the respective products Depending on each of the errors, the scan path deposits the liquid droplets so that printing is performed according to the scan path, and the other of the flat panel device and the respective product. 6. A method according to item 5, forming at least part of the permanent layer for each one.
(Item 7)
Using the printer includes printing according to a scan path that represents a substantially continuous relative motion between the print head and the first substrate, and printing according to the scan path includes: Depending at least in part on the error, the scan path is made to each of the layers for the flat panel device and each of the layers for the other one of the products. Item 3. The method of item 2, wherein liquid droplets are deposited to form at least a portion of each of the layers for the flat panel device and a respective one of the respective products. .
(Item 8)
The criterion is a first visual feature, the stored data is a first stored data, the error is a first error, and the adjusted data is a first Adjusted data, wherein the first substrate comprises visible features at respective locations on the first substrate, the visible features comprising the first visible features;
The method further includes optically detecting each of the visible features and detecting an error between a corresponding location of the feature and at least one expected feature location;
Using the printer is adapted to receive stored data and generate adjusted data for each product that defines where the permanent layer for each of the respective products is printed. Depending on one or more of the errors of each of the respective products depending on the position of the respective products on the substrate with respect to the visual feature. Rendering the stored data defining where the permanent layer for each of the products is to be fabricated according to the one or more of the errors; and adjusting per product Printing the liquid for each of the respective products according to the data
Item 3. The method according to Item 2.
(Item 9)
Rendering comprises, for each of a plurality of processor cores, selecting a respective subset of the printed area of the first substrate, assigning the respective subset to one of the plurality of processor cores, The method of claim 8, wherein the selected one of a plurality of processor cores is used to generate transformed data for each product and is at least partially embodied in a parallel processing environment. Method.
(Item 10)
Selecting, for each one of the respective products, a plurality of processor cores renders the print data depending on errors detected in a manner that is specific to one of the respective products. 10. The method according to item 9, wherein the method is performed in a manner assigned to.
(Item 11)
The stored data comprises a first bitmap having pixels representing print head nozzle firing decisions at respective positions;
Rendering the stored data includes generating a second bitmap that also has pixels representing nozzle firing decisions at respective positions, for each predetermined pixel of the second bitmap,
Identifying at least one pixel of the first bitmap that depends on the error;
Generating a corresponding nozzle decision for the predetermined pixel of the second bitmap that depends on the identified at least one pixel of the first bitmap;
The method of claim 1, wherein using the printer further comprises printing the liquid on the first substrate according to the second bitmap.
(Item 12)
The stored data comprises first grayscale data having pixels, each of the pixels being represented as a multi-bit value representing a desired amount of the liquid at a respective position relative to the first substrate. ,
Rendering the stored data is
Generating second grayscale data having pixels, each of the pixels of the second grayscale data representing a desired amount of the liquid at a respective position relative to the first substrate. Represented as a multi-bit value,
Identifying, for each of the pixels of the second grayscale data, at least one grayscale value of the first grayscale data that depends on the error;
Generating the multi-bit value for each of the pixels of the second grayscale data that depends on the identified at least one grayscale value of the first grayscale data;
The method of claim 1, wherein using the printer further comprises printing the liquid according to the second grayscale data.
(Item 13)
Rendering the stored data further includes converting the multi-bit value of the second grayscale data into droplet density, wherein the droplet density is calculated from the second grayscale data. The step of using the printer selected to produce a volume of the liquid that varies locally according to a gray scale value prints the liquid in response to the flat panel device according to the selected droplet density. 13. The method according to item 12, comprising:
(Item 14)
The stored data defines a predetermined number of raster scans between a print head and the first substrate, and for each respective raster scan, a predetermined cross-scan position of the print head,
The step of using a printer to print the liquid on the first substrate uses the predetermined number of raster scans and associated cross-scan positions, but of the printhead selected depending on the error. 2. A method according to item 1, comprising using individual nozzles.
(Item 15)
The error represents at least one of a tilt, a rotation angle, a distance value, or an amount of scaling, and the step of using the printer is adjusted in a manner adjusted to mitigate the at least one. A method according to item 1, comprising printing the liquid on a substrate.
(Item 16)
Receiving the first substrate includes performing a rough mechanical alignment of the first substrate to the printer, and using the printer includes an expected alignment of the first substrate. The method of claim 1, comprising using the error to perform sub-millimeter fine alignment of the printer data with respect to the software.
(Item 17)
The step of using the printer and the step of treating the liquid to form the permanent layer each comprise printing the liquid and curing the liquid in a common controlled atmosphere. The method of item 1, comprising, wherein said printing and said curing are not collectively interrupted by exposure to an uncontrolled atmosphere.
(Item 18)
Rendering selects a particular processor core among a plurality of processor cores, assigns a transform function and a subset of the printed area of the substrate to the particular processor core, and generates the adjusted data The method of claim 1, wherein the method is at least partially embodied in a parallel processing environment, including using the particular processor core to do.
(Item 19)
Rendering, for each of a plurality of processor cores, selecting a respective subset of the printed area of the substrate and using the aggregated output of the plurality of available processor cores to print a printed image The method of claim 1, wherein the printed image is at least partially embodied in a parallel processing environment having pixels, each representing a nozzle firing decision in a respective region of the substrate. .
(Item 20)
Rendering is further a parameter of a droplet that is expected to be generated from a predetermined one of the nozzles, the droplet volume, the droplet velocity, the droplet trajectory, or the expected droplet Receiving experimental measurement data for each predetermined nozzle of the printhead representing a parameter representing at least one of landing positions;
Depending on the error, assigning individual nozzles of the printhead of the printer to deposit each droplet of the liquid in a particular region of the first substrate may result in the experimental measurement data Done depending on
Depending on the error, assigning individual nozzles of the printer printhead to deposit each droplet of the liquid in a particular region of the first substrate is further dependent on the error. Assigning a firing decision, determining the expected volume of the liquid to be deposited, taking into account the assigned firing decision, and at least one of the expected volume of the liquid to be deposited The method of claim 1, comprising comparing to two thresholds and adjusting the assigned nozzle firing decision if the expected volume violates the at least one threshold.
(Item 21)
The at least one threshold includes a maximum volume of the liquid deposited in the specific area and a minimum volume of the liquid deposited in the specific area, and adjusting the 21. The method of item 20, comprising generating a nozzle firing decision such that an expected volume is not less than the minimum volume and not greater than the maximum volume.
(Item 22)
21. A method according to item 20, wherein the liquid is for forming a photogenerating layer of an organic light emitting diode as the permanent layer of the flat panel device.
(Item 23)
An apparatus for producing a flat panel device,
Means for receiving the first substrate;
Means for printing a liquid on said first substrate, said liquid carrying a material capable of forming a permanent layer for each of the respective products produced in parallel on said first substrate. The flat panel device is one of the respective products and printing is part of an integral printing process for each permanent layer of each of the respective products on the first substrate. Done as a means,
Means for treating the liquid to form a permanent layer of each of the respective products following printing of the liquid;
Subsequent to the processing, means for separating the flat panel device from other respective products on the first substrate,
(A) The apparatus further comprises means for detecting a position of a reference on the first substrate and detecting an error depending on a difference between the position of the reference and an expected feature position. ,
(B) the means for printing generates adjusted data and means for receiving stored data defining where the permanent layer corresponding to the flat panel device is to be printed; Means for rendering the stored data according to the error, and printing on the first substrate is performed according to the adjusted data,
(C) the rendering means further comprises:
In such a manner that the target total volume of the liquid deposited in a particular region is kept within a predetermined error threshold for the desired volume of liquid for the particular region,
An apparatus comprising: means for allocating individual nozzles of the printhead of the printer to deposit a respective droplet of the liquid in the particular region of the first substrate, depending on the error .
(Item 24)
An apparatus for producing a flat panel device,
A machine handler for receiving a first substrate;
A printer for printing a liquid comprising a material capable of forming a permanent layer of each product on the first substrate, wherein the flat panel device is one of the respective products, the printer A printer for printing the permanent layer of the respective product on the first substrate as part of an integrated printing process;
A curing mechanism for curing the material to form the permanent layer;
The flat panel device is separated from the other respective products on the first substrate following the printing of the permanent layer,
The apparatus further detects a position of a reference on the first substrate and detects an error depending on a difference between the position of the reference and an expected feature position; and at least one With a processor,
The at least one processor comprises:
Receiving data defining where the permanent layer corresponding to the flat panel device is to be printed;
Rendering the data according to the error so as to generate adjusted data;
According to the adjusted data, let the printer print on the first substrate,
The at least one processor is further configured such that a target total volume of the liquid deposited in a particular region is reduced to a desired volume of the liquid for the particular region to generate the adjusted data. The printer prints to deposit each droplet of the liquid in the specific area of the first substrate in a manner that is kept within a predetermined error threshold relative to the error. A device that assigns individual nozzles to the head.
(Item 25)
An apparatus comprising instructions stored on a non-transitory machine-readable medium, the instructions comprising: a machine handler that receives a first substrate; and a printer that prints the liquid on the first substrate; Executed by at least one processor of the apparatus comprising: a curing mechanism for curing the liquid to form a permanent layer of a flat panel device; and a detection device for detecting a reference position on the first substrate, the instructions When executed, the machine
Detecting an error depending on a difference between the reference position and an expected feature position;
Receiving data defining where the permanent layer corresponding to the flat panel device is to be printed on the first substrate;
Rendering the data according to the error to generate adjusted data, causing the printer to print on the first substrate according to the adjusted data;
The instructions are further executed such that a target total volume of the liquid deposited in a particular region to produce the adjusted data is a desired amount of the liquid for the particular region. Depending on the error in a manner that is kept within a predetermined error threshold with respect to the volume of the liquid, such that the respective droplets of the liquid are deposited in the specific region of the first substrate. An apparatus causing at least one processor to assign individual nozzles of a print head of the printer.
The present disclosure provides techniques, processes, apparatus, devices, and systems, and products made according to such processes, that can be used to more quickly and more reliably produce products via a printing process. provide.

  The assembly line process uses a printer to print (ie, deposit) liquid droplets on each substrate in a series of substrates. The liquid is then cured, dried, or otherwise processed to form a permanent layer of material (ie, a permanent thin film). Each substrate is used to make one or more products, and a layer will typically be formed for each product carried by the substrate. Once printing is complete, the substrate is advanced, a new substrate is loaded, and the process is repeated. For example, the process for curing the liquid can occur in situ in some embodiments, or at another location within the assembly line, typically printing and processing can be performed with particulate matter, It is performed in a controlled atmosphere to minimize oxygen or moisture contamination. In applications that are considered in detail, printers are used to deposit organic light emitting diode (“OLED”) display panels or a single layer of a solar panel. For example, in the production of high resolution OLED television screens, such a process is used to deposit one or more light generating layers for each pixel of the display (ie, each discrete light generating element). be able to. Although the described techniques can be applied to many types of products other than many types of materials and flat panel displays, the use of printers and related processes can be easily deposited using other processes. It has been found to be particularly useful for depositing layers of organic materials that cannot be done, and therefore many examples presented herein will focus on these materials. As with other types of products and manufacturing processes, printing and layer thickness matching facilitates small, reliable electronic components (eg, micron scale) with low variability and high manufacturing throughput, Must be achieved with a high degree of accuracy.

  As the substrates are moved into and through the printer by different process corners associated with different substrates, assembly line equipment, human handling, and numerous other factors, each substrate is positioned, rotated, scaled. May vary slightly in slope, or other dimensions. Some forms of error may be specific to the substrate (eg, substrate warpage or edge nonlinearity), while other forms of error may represent repetitive system errors (eg, the substrate is the system itself) Are misguided through the printer with repeatable motion errors caused by Since any printing source is intended to print the same product on each substrate in the series, it is generally desirable to detect and mitigate positional errors.

  Thus, in one embodiment, a detection mechanism is used to detect a reference on each substrate when in proximity to the printer. The criterion is used to identify the product position despite substrate-to-substrate errors or other deviations from the expected product position, orientation, and / or dimensions. The detected product position is compared to the expected position and the printer control data is translated or adjusted (in software) (ie how it is rendered to do that data or print) To correct the error in order to precisely align the printed layer with any underlying product dimensions. As used herein, “position” or “product position” generally refers to tilt, scale, orientation, etc., even though these are not individually listed beyond the “position” description. Should be understood to include any of the following.

  It is noted that in a conventional printing process, an assembly comprising one or more printheads, each with a nozzle, is transported against the substrate being printed in one or more raster sweeps. I want. Each sweep defines an “in-scan” dimension of the substrate, and after each sweep, the printhead and / or substrate is repositioned in the “cross-scan” dimension of the substrate in preparation for the next sweep. Each printhead can have hundreds to thousands of nozzles, each of which ejects a droplet of liquid, and the liquid is similar to ink and cured to form the desired permanent layer. Material that will be processed or otherwise processed. For example, using one known technique, the liquid is a monomer or polymer, and following printing, an ultraviolet curing process is used to treat the deposited liquid to form a permanent layer.

  In a more detailed variation of this first embodiment, several mechanisms can be used to harmonize the goals of high speed printing, accurate printing, and precise layer thickness and alignment.

  In a first implementation, once errors (eg, linear or non-linear shifts in position, rotation, tilt, or scaling) are identified, in a manner that depends on the detected deviation from the ideal position, and in the printhead assembly and substrate New nozzle firing decisions can be generated in a manner that does not require a shift or modification of the pre-planned raster sweep between the two. That is, since the printing time (and hence manufacturing throughput) is directly related to the number of raster sweeps required to cover all areas of the substrate that receive the liquid, the number of raster sweeps and their respective The location of is not changed in one embodiment, but rather is a manner that allows printing to occur so as to deposit the intended density and pattern of droplets from the printhead assembly, but any subordinate Nozzle assignments and / or drive parameters (eg, drive waveforms) are converted or rendered to the original printer control data to accurately align with the product layer or to match the intended product geometry In that manner, potentially new nozzles and / or drive parameters are individually assigned. The mode of processing is optional, i.e., in another contemplated embodiment, different offsets of the printhead assembly relative to the substrate (e.g., different advance patterns in the cross-scan direction and / or different numbers of scans or sweeps). Note that the raster sweep is reassessed to potentially employ

  Depending on the environment, simply “shifting” the nozzle firing decision may not necessarily result in the desired deposition parameters. For example, in one considered environment where printing is used to deposit a light generating element in a fluid “well” containing the liquid until it is cured, the error is the nozzle firing quota. May not align cleanly with nozzle spacing or nozzle firing timing so that too many or too few deposited droplets may be located in a pixel well. To address this, in one embodiment, an “anti-aliasing” process is used, and accordingly one or more processors operating under the control of instruction logic are (for the expected liquid volume ( Test the expected position of each pixel well (given the detected error) and if the expected total volume exceeds a threshold amount by more than the ideal value or range of acceptability, the “shifted” nozzle assignment Selectively reconsider and adjust one or more of the number of nozzles that can fire droplets into the pixel well. This then helps to ensure that a precisely intended amount of liquid has been deposited at the intended location.

In another embodiment, the droplets produced by each nozzle are measured experimentally in situ and vary between nozzles or depend on the drive parameters used to eject the droplets from the nozzles. The measured droplet characteristics are stored and taken into account for nozzle assignment. For example, in a first variation, individual nozzles depend on whether the droplet parameters measured in the relevant experiment of the nozzle (and applied drive parameters) meet the criteria for acceptable droplet generation. Can be considered qualifying / disqualifying. Thus, if the position error and associated alignment suggest that an unverified nozzle should not be used for deposition, the instruction logic may indicate other nozzles or different drop ejection parameters (e.g., select The verification data can be used to assign the nozzles to use instead the different drive waveforms for controlling the nozzles that were created. In a second variation, each predetermined nozzle (and each one of a plurality of alternative drive waveforms that may be used to drive the predetermined nozzle) is expected droplet volume, trajectory, and / or Or measured with respect to the drop landing position and then this data is taken into account for error handling or initial nozzle assignment. To cite a simple example, two adjacent nozzles are used to deposit 9.97-10.03 picoliter (pL) droplets respectively into fluid wells for a total volume of 20.00 pL of liquid. If it is relied upon and the substrate error suggests that two other nozzles will be used to perform the deposition, (a) non-expected to produce 9.90-10.10 pL droplets Adjacent nozzles can be assigned to perform deposition (with the expected value of 20.00 pL total volume) and / or (b) expected in the fluid wells despite detected errors. So that the nozzle drive details (e.g., drive waveform) of one or more nozzles produce droplets with adjusted volume or trajectory characteristics from preselected nozzles so as to maintain the total volume It can be section. These types of processing are also optional and may not be required for all embodiments.

  Note that the printer control data can take many forms, depending on the embodiment. In one implementation, “recipe” information describing the desired dimensions (eg, including thickness) of the desired layer for each product is stored as a cached template and then adapted to run-time errors. Can be rendered in an adjustable style. In a second implementation, this recipe information is partially preprocessed and pre-stored in an object (vector) or other representation as a cached template for use in processing each substrate as well. Can be. As position or alignment errors are detected for each new substrate, the template is retrieved and corrected accordingly when rendering the final printer control data (ie, nozzle firing decision, raster sweep, and associated timing, as appropriate). Is done. To quote another example, in another implementation, recipe data is received for a product (and / or array) and rendered into a bitmap that presents a print decision. A bitmap is essentially a nozzle firing at each node of the printing grid that represents a trigger that would cause the nozzle to fire or not fire a droplet at discrete locations on the substrate during the scan movement of the print head relative to the substrate. An array of decisions or equivalent data. This information can also include variable nozzle-by-nozzle drive waveform definitions or selections. The bitmap is cached as a template, read out at runtime, modified to distort the print to match the error via direct processing on the bitmap, and then used to control the print The Inevitably, other embodiments exist, i.e., in each variation, the detected error is in some form to customize the print in consideration of the detected error to achieve layer alignment. To take into account the rendering process.

  In each embodiment, the substrate (or any individual product expressed thereon) in any given manufacturing iteration is detected to facilitate highly accurate printing without increasing printing time. Mismatches can be corrected in software at least partially in real time. The correction then reduces the need for time consuming very high precision mechanical alignment or relocation, as well as the need for very high precision alignment mechanisms. This reduces cost and increases manufacturing throughput without sacrificing accuracy and reliability. In one detail-considered application related to the arrayed printing of multiple large OLED TV screens, a single large substrate is used for 6-8 HDTV screens (e.g., used interchangeably). Panel or sub-panel). It is now considered that for a successful manufacturing process, the layer printed on the substrate should take only 90 seconds per substrate. In some embodiments, this maximum printing time is expected to be 45 seconds or less. Since each screen or panel involves millions of pixels, the alignment should be precise for the production of high quality displays. Accordingly, it is desirable to print a liquid on a substrate using a process that takes less than a few seconds (eg, 2 seconds) to detect and correct substrate and / or product misalignment or other errors. . The disclosed embodiments facilitate the manufacturing speed (and manufacturing throughput) and help produce a smaller and more precise alignment product.

  Many manufacturing applications are relatively robust against the coarse mechanical alignment of the product being assembled. With respect to the disclosed embodiments, the fine precision performed by one or more processors (ie, machines) operating under the control of software (instructions or instruction logic stored on a non-transitory machine-readable medium). Alignment typically adjusts for position offset, rotational offset, tilt, scaling error in millimeters to less than a millimeter (eg, nanoscale to hundreds of microns or less), or other distortion. For some applications featuring very high density precision structures (eg, HDTV applications with millions of pixels) and / or involving thousands of printhead nozzles, error handling can be significant Note that computational resources may be required. As noted above, in one embodiment, it is desirable to perform error compensation using hardware and / or software logic in a few seconds, eg, 2 seconds or less. To facilitate this goal, certain embodiments below are for hardware designs that rely on parallel processing (eg, multiple processors or multi-core processing) and process thread allocation that allows this computation to occur quickly. Present the technique. For example, foreseeing an embodiment that will be discussed further below, a supervisory processor or group of processors detects errors and renders or transforms the original recipe information (or other cached template data). The formulation for can be determined. The formulation can be linear across the substrate or can vary locally (eg, non-linear or non-continuous). The supervisory processor then assigns a geographic partition and associated affine transformation to each core of the multi-core processor for processing. In one design, each core has its own dedicated memory for data processing and manipulation, and the supervisory processor identifies the geographic partition that is modified by each core and associated modification algorithm (eg, affine transformation). Providing this information to each core, and then each of the plurality of cores is assigned to contribute to ending the converted output, representing the adjusted printer control data for the board as a whole. Perform the task. The converted output is suitable for use in immediate printing. If desired, the memory used for the operation is also designed to provide direct memory access (DMA) to accelerate the operation of the cached template and its adaptation to printing and use in printing. You can also. For embodiments that use tens to hundreds of cores or more, this substantially reduces the processing time required to re-render the desired print job. Other designs are possible. For example, instead of assigning different terrain to each core, different mathematical operations (or different sequential processes) can be assigned to each core. As this example should be clear, almost any partition of function or processing can be provided to maximize efficiency or otherwise reduce processing time. A parallel processing environment as discussed above should also be considered optional for the other techniques described herein.

  The present disclosure describes several hardware (ie, circuit) implementations that facilitate highly accurate printing in an assembly line process, and software techniques that can be used in place of or in addition to such hardware And provide. Generally speaking, the features discussed below can be mixed and matched (or otherwise optional) and can vary according to implementation. For example, one embodiment discussed below provides several (eg, 16) customizable nozzle drive waveforms for each of the many nozzles (eg, hundreds to tens of thousands) used by the printer. Hardware that can be used to pre-store Each waveform is pre-selected to produce slightly different expected drop parameters (eg, volume, trajectory, etc.) that provide a series of possible drops that can be generated from that nozzle. The system pre-selects these waveforms to provide a series of selections and pre-programs the various waveform selections to drive each nozzle circuit. Then, at runtime, the system simply selects one of the waveforms. In one embodiment, there are 16 such choices (eg, one “zero” waveform representing a non-firing decision, and 15 variable drive waveforms). In another embodiment, the default waveform can be programmed asynchronously (ie, pre-), and then “rise” so that which waveform is most recently programmed as the default. The value “trigger” can be applied. Once again, these various features are optional and not required for all embodiments, and the various disclosed features may be used in any desired combination or permutation suitable for implementation. May be. All such combinations or permutations, and any such combinations or permutations are contemplated by the teachings of this disclosure.

  An implementation that is considered in detail can include an apparatus comprising instructions stored on a non-transitory machine-readable medium. Such instruction logic entails the tasks described on the input operands depending on the instruction that takes a particular action or otherwise produces a particular output when the instruction is finally executed. Written in a manner having a certain structure (architectural feature) to make one or more general purpose machines (eg, a processor, computer, or other machine) behave as a general purpose machine having a structure Or can be designed. As used herein, a “non-transitory machine-readable medium” is not limited, regardless of how the data on the medium is stored, but the instructions are later read by a machine. Any tangible (ie, physical) storage medium that can be issued, including random access memory, hard disk memory, optical memory, floppy disk or CD, server storage, volatile memory, and other tangible mechanisms Means. A machine-readable medium may be in a stand-alone form (eg, a program disk or a solid state device), or a larger mechanism such as a laptop computer, portable device, server, network, printer, or one or more thereof. It can be embodied as part of another set of devices. The instructions are in different forms, for example as Java code or script, as code written in a particular programming language, as metadata that is effective to invoke an action when invoked ( (E.g., as C ++ code), as a processor specific instruction set, or in some other form. The instructions can also be executed by the same processor or different processors or processor cores, depending on the embodiment. Throughout this disclosure, any can generally be implemented as instructions stored on a non-transitory machine-readable medium, such as microelectronics, microoptics, “3D printing”, or others Various processes that can be used to fabricate products using the current printing process will be described. Depending on the product design, such products may be in a sealable form or other printing, curing, manufacturing that may ultimately produce a finished product for sale, distribution, export, or import Or as a preliminary step for other processing steps. Also, depending on the implementation, the instructions can be executed by a single computer, in other cases using, for example, one or more servers, web clients, or application specific devices. Can be stored and / or executed on a distributed basis. Each function described with reference to the various figures herein is either on a single media representation (eg, a single floppy disk) or on multiple separate storage devices. Can be implemented as part of a composite program or as a stand-alone module. The same applies to print images or printer control data generated according to the process described herein, i.e., the recipe information, template, or the result of processing such recipe information or template is the same machine. It can be stored on a non-transitory machine-readable medium for temporary or permanent use, either above or for use on one or more other machines. For example, printer control data is generated using a first machine and then transferred to a printer or manufacturing device, eg, for download over the Internet (or another network), or on another machine Can be stored for manual transport (eg, via a transport medium such as a DVD).

  Also, in the above, reference is made to the detection mechanism and the criteria recognized on each substrate. In many embodiments, the detection mechanism is an optical detection mechanism that uses a sensor array (eg, a camera) to detect a recognizable shape or pattern on the substrate. Other embodiments are not premise on the array, for example, a line sensor is used to sense the reference when the substrate is loaded into the printer or advanced in the printer. Can do. While some embodiments rely on dedicated patterns (eg, special alignment marks), other embodiments rely on recognizable substrate features (including any previously deposited layer geometry). Rely on and note that each of these is a “standard”. In addition to using visible light, other embodiments can rely on detection of ultraviolet or other invisible light, magnetic, high frequency, or other forms of substrate details relative to the expected print location.

FIG. 1A shows the layout of the substrate, which is intended to accept a liquid print that can form a permanent layer of the respective product at location 107, with the overall print area intended for access by the nozzles of the printhead. , Represented by a dashed box 103. FIG. 1B is similar to the dashed box 103 of FIG. 1A, but without correction, the substrate is unintentionally offset in place so that printing occurs in the wrong place, resulting in misalignment. A print area 125 is shown. FIG. 1C shows a print area 135 similar to the dashed box 103 of FIG. 1A, but with substrate rotation errors. 1D is similar to the dashed box 103 of FIG. 1A, but the substrate misalignment represents a scaling error, for example, each printed product may have a dimensional error (as well as a positional error). 147 is shown. FIG. 1E shows a printed area 155 similar to the dashed box 103 of FIG. 1A, but where substrate misalignment (eg, edge distortion in an edge-guided alignment process) results in distortion in two independent dimensions. FIG. 1F shows a printed area 163 that is similar to the dashed box 103 of FIG. 1A, but with different types of errors each affecting a different part of the substrate. FIG. 2A provides a flowchart 201 for alignment of one or more printed deposition areas with respect to a substrate position. FIG. 2B provides another flowchart 221 related to the alignment of one or more printed deposition areas with respect to the substrate position. FIG. 2C is a screen shot from the recipe editor and panel definition software used to define the layer geometry for each product design in the array (ie, the array uses a common substrate). The layer deposition is repeated for a series of substrates). FIG. 3A provides a flowchart 301 for one particular method of adjusting the “template” to fine tune the print depending on the detected error. FIG. 3B provides another flowchart 351 for one particular method of adjusting the “template” to fine tune the print depending on the detected error. 4A-4I are used to discuss the handling of alignment errors, and more particularly to illustrate the process for assigning or reassigning printhead nozzle firing decisions to fine tune the print. The FIG. 4A shows that the presence or absence of an “X” at a location in the “print grid” indicates that a drop is fired by a printhead in a particular area (eg, from nozzle 410 at position 404), respectively. Or shows the layout of printhead 407 relative to region 403, which represents that it will be retained in a particular region (eg, from nozzle 410 as it traverses position 405). For example, FIG. 4A may be a section of a “bitmap” (ie, representing an intended nozzle firing decision over a portion of the substrate as the printhead and substrate move relative to each other in a scanning motion). . Note that as used herein, the term “bitmap” refers to the data for each nozzle, regardless of whether each nozzle's data consists of a single bit or multiple bits. . For example, in one embodiment, 4 bits of data can be used (i.e., one of which represents a decision not to fire a nozzle and the other can be "sent" to the nozzle, a different reprogramming Represents 16 possible values, representing the nozzle drive waveform generated). Clearly, many alternatives are possible. FIG. 4B shows the layout of the original print grid 403 from FIG. 4A, but the substrate is misaligned with respect to the intended print area, and different nozzles make this alignment error without adjusting the scan path. Can be used to compensate. In the depicted example, misalignment may not align “clean” with nozzle spacing or “pitch”, eg, more or less nozzles as the printhead and substrate are moved relative to each other. Note that different sets of printhead nozzles may overlap the desired print area 412. FIG. 4C shows a layout 421 where the print grid is seen overlaid on the desired print area 412 from FIG. 4B. FIG. 4D shows the relative layout 431 of the region 403 from FIG. 4A, but the intended printed region or product geometry 433 has been rotated relative to that region due to substrate errors or other misalignments. It can be seen. FIG. 4E shows a layout 441 in which the print grid is seen overlaid on the desired print area 433. FIG. 4F shows the original printer grid 403 layout, but for a particular substrate, there is a detected scaling error as reflected by the desired print area 455. FIG. 4G shows a layout 461 in which the print grid is seen overlaid on the desired print area 455 from FIG. 4F. FIG. 4H shows the layout of the original printed grid from FIG. 4A, but there is a detected tilt error 475, which is actually (shape due to some type of distortion or linear advance error affecting the substrate, for example. It is desired to print along the outline of the parallelogram (represented by 475). FIG. 4I shows a layout 481 in which the print grid is seen overlaid on the desired print area 475 and is presented herein with print corrected using the print grid of FIG. 4I. Using the technique, the print can be precisely aligned (ie, represented by shape 475). FIG. 5A shows a circuit used to assign or adjust nozzle drive waveforms (ie, electronic drive signals) to different nozzles of the printhead. FIG. 5B shows the corresponding prints where the stored data allows programmable waveform definitions (and thus variable volume, droplet trajectory, droplet landing location, velocity, or other droplet parameters per nozzle). Fig. 4 shows a drive circuit for each nozzle that can be used to store head nozzle data. FIG. 5C shows an exemplary waveform and is used to help explain how the waveform can be used to vary nozzle droplet parameters with programmable adjustments. FIG. 5D shows another embodiment of a drive circuit for the printhead nozzles. FIG. 6A is an illustrative diagram illustrating a series of optional layers, products, or services that can each independently implement the techniques introduced herein, eg, as described herein. The technique presented can be in the form of software (by number 603) or as printer control data (by number 607, used to control the printer to print on the substrate) or (numbers 613, 615). Or as implemented by products made on the basis of these techniques (as exemplified by 617). FIG. 6B provides a schematic diagram of a fabrication device that includes a printer. FIG. 6C shows a plan view of the printing area in the production device, which can optionally be contained in the gas enclosure, i.e. so that printing takes place in a controlled atmosphere. FIG. 6D provides a functional block diagram of a fabrication device that includes a printer. FIG. 7 provides a flow chart 701 that is used to measure the effects of individual nozzle variations and / or nozzle drive waveform variations, and related compensation techniques. In one embodiment, not only is printing adjusted for position errors, but in addition, nozzle-to-nozzle variations are taken into account so that printing occurs within carefully defined tolerances. FIG. 8A is an illustrative schematic showing a droplet measurement system capable of measuring the droplet volume of each nozzle of a large printhead assembly. For example, such a system can be used to measure inter-nozzle variations referred to above. FIG. 8B is a schematic diagram of a method showing various processes and options associated with measuring the droplet details of each nozzle to achieve a reliable understanding of expected droplet characteristics. FIG. 8C shows a flow diagram associated with the drop measurement embodiment. FIG. 8D shows a flow diagram associated with nozzle verification, that is, in one embodiment, measurements are used to consider a nozzle qualifying or ineligible (in addition to other purposes described herein). Can be done. FIG. 9A provides a block diagram associated with the raster process. More specifically, FIG. 9A can be integrated with the substrate error (eg, position, rotation, scaling, or tilt error) how the droplet measurement results were detected to provide precise printing. It helps to explain. The shaded area (907) represents a single scan path while the transparent area (908) represents another path. FIG. 9B shows a flow diagram associated with printing on a substrate as part of the manufacturing process. FIG. 9C shows another flow diagram associated with printing on a substrate as part of the manufacturing process. FIG. 10A shows a flow diagram associated with printing on a substrate as part of the manufacturing process. FIG. 10B shows another flow diagram associated with printing on a substrate as part of the manufacturing process. FIG. 10C tests to see if print acceptance, eg, “conditioned” printer control data, still meets one or more desired criteria (eg, th ₁ ≦ volume ≦ th ₂ ), A flow chart 1041 for terrain anti-aliasing is provided that further adjusts printer control data if one or more desired criteria are not met. FIG. 10D provides a flowchart 1061 relating to incorporating updated nozzle droplet data into the rendering of printer control data to reduce substrate printer variation. FIG. 10E provides yet another flowchart 1081 relating to incorporating updated nozzle droplet data into the rendering of printer control data to mitigate substrate printer variation. FIG. 11A is a block diagram illustrating a parallel processing environment. FIG. 11B is a block diagram illustrating how print image rendering and / or rasterization occurs in a parallel processing environment.

  The subject matter defined by the following claims can be better understood with reference to the following detailed description, which should be read in conjunction with the accompanying drawings. This description of one or more specific embodiments, which are set out below to enable the construction and use of various implementations of the technology described by the claims, is listed below. Are intended to illustrate their use. Without limiting the foregoing, the present disclosure may be used to fabricate a thin film for each of a plurality of products (or other arrays of products) of a substrate as part of an integrated repeatable printing process. Several different embodiments of the technique are provided. Various techniques are available as software for performing these techniques, such as control data (eg, printing) for forming such film layers in the form of a computer, printer, or other device that executes such software. Image), as a deposition mechanism, or in the form of an electronic or other device fabricated as a result of these techniques (eg, having one or more layers produced according to the techniques described) Can be embodied. Although specific embodiments have been presented, the principles described herein may also be applied to other methods, devices, and systems.

(Detailed explanation)
Having thus described several basic embodiments, the present disclosure will now be directed to a more detailed implementation. 1A-1F are used to illustrate various principles associated with printer control data adjustment in a repeatable manufacturing process.

  FIG. 1A illustrates the layout of the hypothetical substrate 101. More specifically, the substrate already has a number of alignment marks 105, a printed area 103 defined for the alignment marks, and a base layer formed or deposited as part of the product formation process. And an area 107 corresponding to some products. For illustrative purposes, the substrate 101 will be used to form a number of products 107 and each “panel” that will be cut from the substrate to form an independent solar panel or display device. It can be assumed that the present invention is not so limited. In one embodiment, each panel 107 is an independent OLED panel and the substrate 101 is a large piece of glass.

  Despite the presence of multiple products, it is desirable to employ a printer to eject liquid onto the array or substrate 101 in an integrated printing process. The liquid has a well-planned thickness so that the material and / or liquid becomes a permanent part of each resulting product, following the deposition of the liquid and its curing or other treatment So as to carry the material. In an optional embodiment, the layer thickness is applied using controlled droplet deposition, where the volume or density of deposition liquid per unit area is used to build the layer thickness. That is, the liquid will have a limited diffusion, resulting in a blanket liquid coating without undesirable holes or gaps, or the liquid will otherwise provide all the planned thicknesses in different ways. In a brazing manner, it is deposited or hardened in a manner that would be geometrically confined. Even if the deposition fluid is typically colorless and is not used to create any type of shade (ie, “half”, “blend”, or otherwise), the process generally involves In this specification, it is referred to as “half-tonig”. In a typical implementation, the printer prints the entire substrate (ie, the layer of each product in the array), and then the substrate is transported from the printer to a separate curing chamber where all controlled atmospheres (eg, liquid In the presence of moisture, oxygen, or other forms of unwanted particulate matter that prevents exposure to nitrogen or other ambient non-air atmosphere). Note that there will be. For this example, it will be assumed that recipe data is generated in advance that asks the printer to deposit the desired material at discrete locations within the boundaries of the overall print area 103. Briefly, the recipe data for each product describes any desired details such as the layer thickness and dimensions of that product, as well as corner rounding or edge accumulation profiles, and the recipe data for the array Describe where the layer (ie, product recipe data) is located relative to the substrate. Recipe data, when processed or rendered, may instruct the inkjet printer to deposit liquid into each individual product area 107 on a reproducible basis that will be repeated for many secured similar substrates. . Recipe data (or any processed version thereof) is stored in memory for use as a template, which may be used to deposit liquid on the printer on each substrate in a series of substrates. “Cached”. FIG. 1A shows one of these substrates (with many products or “panels” thereon) that represents an ideal deposition on the substrate.

  As noted above, unfortunately, in practice, each array or substrate will affect the alignment of the deposited layer for one or more of the products, either in its structure or in its position. There may be uniformity, rotation, scale, tilt, or other alignment issues. For many manufacturing processes, such distortions can be tolerated, but for applications such as microelectronics, or where very precise alignment is otherwise required, such alignment problems can reduce feature size. Limiting, creating product defects, or otherwise increasing the time and / or cost of manufacture. It is desirable to avoid and / or compensate for these problems and to provide printing that is automatic and as fast as possible while maintaining accuracy.

  To do this, in one embodiment, hardware logic and / or instruction logic ideally does not require adjusting the planned scan path of the printer (eg, only nozzle data is adjusted). ) Adjust the nozzle firing data to take into account the detected error. This is not required for all embodiments. A fiducial (eg, a substrate alignment mark or optically recognized feature) is detected optically as each new substrate is loaded, advanced, or positioned in the printer. The associated image data is then used by the processor to compare the actual position of the substrate (or its panel) with the expected position, and depending on the deviation, the printed image is printed with panel layout information (and optional Adjusted to align with the previously deposited layer). The alignment corrections made using these techniques are typically micron-scale or finer (e.g., position errors that are less than a millimeter, in some embodiments 100 microns or even substantially more Less than position error).

  FIG. 1B shows a situation where the substrate is actually misaligned with the inkjet printer. In this case, whether or not due to edge errors, mechanical positioning errors, substrate process corners, or other problems, this misalignment creates the possibility of alignment errors between the layers of each product being fabricated. Note that the depicted errors are exaggerated in scale for ease of illustration. More specifically, an xy Cartesian error (represented by the number 123) will offset individual product layers (107) relative to the ideal print area 125 on the substrate if left uncorrected. This offset can cause problems for any subordinate (or later deposited or fabricated) product layer. The techniques disclosed herein may shift the ink deposition in a manner that matches the alignment error, eg, the printing process is offset (or otherwise) in a manner that matches the substrate and / or panel position error. Used to adjust the printer control data as This detected error is corrected by generating or adjusting printer nozzle control data in a manner consistent with such error. As described above, if the recipe data is preprocessed, an optional technique can be used to reassign any previous nozzle firing pattern in an error dependent manner. In one embodiment, this reallocation typically involves more than simply shifting all nozzle firing decisions by a vector amount, i.e., to retain precise liquid filling for the desired area. To adjust for other factors (e.g., anti-aliasing as discussed further below), taking into account nozzle-to-nozzle variations and misalignment with the deposition area of interest (e.g., pixel well) and printed grid, Note that the nozzle firing decision can be reassessed. In one embodiment, the reallocation depends on print grid points associated with any preprocessed or rendered data (eg, relying on a template bitmap) and / or nozzle drive parameters Is effectively performed by a process operating at the bitmap or vector representation level. The use of other adjustment techniques such as scan path variation (eg, adjusting printhead offset and rasterization changes) is also possible.

  FIG. 1C represents a situation where the substrate has another form of error represented by the rotation angle α (number 133). If left uncorrected, this error will potentially cause individual product layers (107) to potentially misalign with other product layers (eg, relative to the desired printed area 135 of the substrate). Using the techniques discussed herein, this error is detected by misalignment, for example by taking into account rotation adjustments in the rendering of printer data in hardware, software, or a combination thereof. Can be corrected. Once again, if a template print image, such as a bitmap, has already been generated, this correction is optionally performed directly from the preprocessed bitmap (for example, without the need to generate a new print image from the recipe data). By calculating nozzle firing assignments and / or drive parameters, it can be done directly on a copy of the image. As in the previous example, adjustment results in a per-substrate printing process where the deposited layer exactly matches the intended product geometry, despite the error, and printing is done almost immediately. Can do.

  Note that it is an implementation decision whether the error is considered in the preprocessed data (such as a template or bitmap) or applied directly to render the printer data first from the recipe information. . For OLED devices with millions of pixels (and millions of potential droplet deposition points per substrate), the per-board adaptation of pre-rendered bitmaps is computationally dependent on the hardware capabilities supported. Can be very expensive. In some embodiments, it may be faster to render the printer data directly from the recipe data in a manner that takes into account the detected registration error; in other embodiments, the bitmap (or other previous The use of (treatment) can be more efficient. A number of processing options are available for embodiments that further use the parallel processing features provided below.

FIG. 1D represents yet another type of error, depicted in this case as being linear across the substrate or product array, representing a scaling problem. For the purposes of this example, the substrate should be assumed to be slightly larger or smaller than intended, eg, in such a way that the error is at least locally uniform or linear due to temperature changes. The substrate may be slightly shrunk or expanded. Thus, providing an example of a simplified, with respect to the hypothesis, all dimensions also placed in the template image, k ₁ and k ₂ are scalars, X and Y represent Cartesian coordinates on the substrate, k ₁ It is assumed that it should be adjusted by X, k ₂ Y. Note that the situation where the scaling error is in only one dimension (eg, the X dimension) presents a simplified example where k ₂ = 1. If left uncorrected, the scaling error will result in individual product layers (107) that are slightly smaller (149) than the offset error as well as the desired footprint (148) for each panel. Similarly, the overall print area is seen slightly smaller (145) than the desired print area (147) of the substrate. To correct this error, the scaling is corrected according to k ₁ and k ₂ . Once again, in one embodiment, this is converted printer control (eg, using potentially different nozzle firing decisions and mappings, either directly from recipe data or from preprocessed template data). Can be implemented by taking into account the variables k ₁ and k ₂ to generate data. Anti-aliasing and / or nozzle droplet details (as discussed below) are also optionally intended for the local density of the deposited liquid at a given location or unit area (ie thickness can be controlled). It can be applied / considered as part of this transformation to ensure that it corresponds to the layer dimensions. As in the previous example, the adjustment is made on a per-substrate basis, where the deposition layer accurately matches the intended product geometry, and the desired deposition fill, density, and volume are correct, despite the substrate position error. Brings the printing process. Certain adjustments to FIG. 1D can result in substantial changes to scan nozzle assignments per printhead, potentially affecting the number of scans (depending on the nature of the embodiment and error), depending on the embodiment. Please note that That is, in one embodiment, as previously referenced, nozzle firing decisions are reassigned depending on the techniques disclosed herein (eg, without changing the position or number of scans). In another embodiment, the scan path can be re-optimized taking into account droplet data and errors per nozzle, and a different scan path and a greater or lesser number of scans can be used. Regardless of the embodiment, printing can be done almost immediately as planned, and substrate error independently considers the position error of each substrate (or part of a given substrate on a local basis). Compensated by software, hardware, or both, depending on the dynamic adjustments made.

  Once again, all errors depicted in these figures are visually exaggerated for illustrative purposes, i.e., the scale where the arrayed area (e.g., glass substrate) is several meters wide by several meters long Note that the coarse mechanical alignment is typically accurate to a few millimeters or better. Thus, in many embodiments, the disclosed techniques are used to correct relatively small misalignments (eg, tens to hundreds of microns or even less).

  FIG. 1E represents yet another type of error where, for example, individual product areas or substrates generally undergo a tilt. Here, the error is considered to result in a Y position error (horizontal with respect to the page of the drawing) that varies depending on the X position (vertical direction with respect to the page of the drawing) on the substrate. As an example, this type of error can be caused by substrate edge errors or by transport errors that uniquely affect a particular printer or manufacturing device. In the figure, all dimensions and positions are assumed to be tilted by X ′ = X, Y ′ = fn {X}, where X ′ and Y ′ are adjusted coordinates as represented by graphic 153. It is. Once again, note that the tilt situation will also in practice also result in a position offset error for each panel 107 that is added to the panel tilt error (this is one row for the next row in the figure). Considered horizontal offset of the middle panel). If left uncorrected, tilt errors will deform individual product layers (107) for the desired footprint. To correct for this error, the printed image is once again adjusted / converted to pre-tilted panel printing by software to match the underlying error. As with the previous embodiment, the conversion performed can involve more than simply shifting the nozzle firing decision, eg, ensuring the deposition of the appropriate volume of ink in the appropriate target area. Additional processes can be featured. This optionally results in substantial changes to the scan nozzle assignment per printhead, as previously discussed, and can potentially affect the number of scans. Printing is once again almost instantaneously and substrate errors are corrected in software by real-time adjustment of the template print image or in the rendering of recipe information for a particular substrate.

  FIG. 1F represents a hypothetical case in which the described errors can affect each panel independently of each other, and in addition to or instead of such errors. For example, FIG. 1F shows that the first panel 107 ′ can be seen tilted with respect to the desired print area 163, while the second, third, and fourth panels (107 ″, 107 ″ ′, and 107 "") depict cases that are each considered to have rotation, offset, and scaling errors relative to the other panel 107. Note that the 107 "" scaling error only affects this panel, so it is schematically depicted as having a small or no position offset. Applying the teachings introduced herein, each panel error is corrected separately by adjusting the rendering that would be used to print a given array or substrate. In this regard, the error of any given panel or part thereof can be complex and a superposition or combination of transformation techniques for one or more of position offset, rotation, tilt, and scaling errors. The correction is modeled as: This can be done on an independent basis for a group of panels, a single panel, a space spanning portions of multiple panels, or a portion of any given panel. For embodiments that cache a pre-rendered template print image, for each new board (eg, every 45 seconds), a copy of the cached image is required without the need to mechanically relocate that particular board. Can be re-assigned and fired to reduce alignment, orientation, or other errors. For each subsequent board, for example, every 45 seconds, an instance of cached template printer control data can be loaded once again and used to correct positioning or alignment errors specific to the new board. Generate rendered data that can be immediately applied to printing.

  There are several techniques for fine alignment of printer control data to match substrate or panel deviations from expected positions. First, one contemplated embodiment superimposes the printed grid on the actual (ie, detected) substrate and / or panel location. As described above, the printing grid is a “horizontally separated” node representing the nozzle pitch on the printhead and, for example, “1” as the printhead and substrate are “scanned” relative to each other. It can feature "vertically separated" nodes representing digital timing at which the nozzle can be fired at a "per micron" distance. For each node in the overlapped space, the system (ie, instruction logic running on one or more processors) fires at that point so that each nozzle deposits a liquid droplet. Determine if it should be done. With respect to the superimposed space, the firing decision at each node is a function of the original “recipe” data of the ideal substrate as well as the deviation of that node's position from the equivalent point in the perfectly matched substrate. In one implementation, printer control data is “pre-rendered” to form a bitmap (ie, nozzle firing decisions have already been assigned in a manner that assumes uniform alignment). Each firing decision of the printed grid superimposed with the panel position can be mapped to an equivalent point in the original bitmap via transformation. To provide an example, if a given substrate is offset by 31.0 microns “on the right” where it should be, it first identifies the same node location in the original bitmap. Then identify each point in the printed grid by identifying the point 31.0 microns “on the left” of that location, and finally identifying the firing details associated with the node closest to that shifted point The launch decision can be calculated. As will be discussed further below, in particular, where liquid is to be printed into the pixel wells, in such embodiments, such a process may be performed with too much or too little liquid. Drops can be deposited within the well boundaries, and adjustment processes such as anti-aliasing (discussed below) can be employed to mitigate this possibility. In the second implementation, each firing decision in the superimposed grid is a weighted function of the firing decision from a pre-rendered template (eg, a point 31.0 microns offset “to the left” in the original bitmap). The weighted average of the firing decisions of the four printing nodes closest to. Naturally, many variations and possible adjustment mechanisms will occur to those skilled in the art. Other embodiments may employ in-situ measurements of imported alignment data or nozzle parameters, which are then optionally rendered for printer control data for a given substrate, anti-aliased, drop density ( Halftone pattern) adjustment, and other process considerations. Each of these is more reliable by using one or more additional processes to ensure that the proper amount and / or density of liquid is printed in discrete regions of the substrate. Helps create deposits. In some embodiments, the number of scan paths or printhead cross scan offsets can be reconsidered to optimize printing time. Various combinations of the disclosed techniques and their relative benefits will be apparent to those skilled in the art in view of the following discussion.

  FIG. 2A is used to introduce some exemplary processing tasks. More particularly, FIG. 2A shows that certain functions are performed as part of an offline process (ie, above the dashed line 209) or during run time (ie, below the dashed line 209). Fig. 2 shows a flowchart 201 drawn as done (online as part). A computer (eg, a stand-alone or part of a manufacturing device such as an industrial or manufacturing printer) first receives information describing the desired deposition of a layer across the substrate by numeral 203. Several variations are contemplated as described on the right side of FIG. 2A. For example, one embodiment (204) allows for the production of many products at a time as part of an ordered deposition. Returning briefly to FIG. 1A, for a particular manufacturing process, the layers for each of the 15 panels shown may be exactly the same in width, length, and thickness as one another. For example, a hard recipe such that a single recipe (representing one of these panels) generates a total print image with total print data of expected print locations for 15 individual products. It can be electronically loaded and arranged by ware logic or instruction logic or both. This need not be true for all embodiments. For example, both on the same substrate, a first size TV screen can be represented by a first horizontal row of panels, and a second size TV screen can be represented by a second row of panels. Each panel may have different manufacturing details in this example. Regardless of whether multiple products are represented by each substrate, the method can be applied to the desired location of each layer (e.g., "upper left" corner location and length and width dimensions, as per numeral 205, desired as per numeral 206). Receiving information such as layer thickness). In one example, “recipe editor” software can be used to define layer parameters for a substrate carrying one or more products (see, for example, the discussion of FIG. 2C below). The computer then stores information representing the desired layer for use as a template when processing a number of substrates, ie, each in sequence execution as part of an assembly line process. The stored information is optionally a bitmap or other so as to represent a print of an “ideal” substrate (ie, all products are perfectly positioned on them) Pre-rendered as a form or otherwise somehow associated with a default substrate / product location.

  During run time, as each new substrate is loaded into the printer or otherwise received, the substrate is roughly positioned using mechanical means. For example, this step can be performed by the number 211, via a machine handler, or via an edge guide that “roughly” positions the substrate in the desired printing position. The printer then detects the actual substrate position using the precision detection system with the numeral 213. In one embodiment, this is done using a high precision camera that images a region of a reference (a known pattern in or on the substrate), the camera is precisely positioned relative to the printing system, and position detection software Are used to perform image processing to identify the exact location of the reference, for example, down to the micron. Advantageously, the reference may be a two-dimensional pattern or set of patterns that allows determination of the rotational orientation and / or scale and / or tilt of the substrate. The more complex the criteria (or the greater the number of independent features that can be recognized), the greater the accuracy and / or number and / or complexity of the alignment or location problems that can be corrected. Providing an example of this, one embodiment (see FIG. 9A, also discussed below) is that at least two dedicated panels allow all panels on the substrate to detect or reduce errors per panel on a panel basis. Consider that it may have alignment marks. This is not required for all embodiments. With either detection mechanism, the computer (and / or one of its processors) then (1) the detected panel position, dimension, and orientation, and (2) the expected panel position, Deviations between dimensions and orientations are derived and the print job is adapted (215) as necessary to align the print with any lower layers and / or product geometries. With numeral 217, the panel is then unloaded from the printer and optionally cured with numeral 218. A new substrate can then be loaded for a new print job, as indicated by reference numeral 219.

  As an example representing a hypothetical printing process for OLED or solar panel manufacturing, the typical layer thickness of the panel depends on the particular layer being “printed” (and then cured or otherwise processed). From about less than a micron to several hundreds of microns. The array definition information can be used to determine the location according to the recipe of the particular panel received and linked to the print location of each panel. For example, in a configuration with 15 panels per board, the first panel “recipe” is replicated five times in each of the first two rows of panels on the board (ie, each at a precise location). Yes, the second unique panel recipe was replicated five times in the third row of the panel (ie, in a precise location, assuming a second recipe different from the first recipe) As shown, array definition information may be received. The array definition may identify each panel location and the particular recipe used for that location in this example. Note that the example of a homogeneous continuous layer for any given product is merely illustrative. In some embodiments, the layers may be patterned within each product (as opposed to being a homogeneous coating, such as a thin film encapsulation for a particular product or panel), or the thickness may be Can be varied within a layer as defined by a particular panel recipe. Continuing with the three-row example of 15 panels, the third row of products is potentially used to “build” a greater thickness than is created in the first two rows of the panel. Different layer thicknesses can be featured, with higher density halftoning.

  FIG. 2B provides a flowchart 221 of an exemplary printing process for each substrate in a series of substrates. More specifically, as each new substrate is received in the series (223), the machine handler transports the substrate to a first location (225). This positions the substrate in such a way that the expected reference position is within the camera's field of view. In one embodiment, numeral 227 causes one or more cameras to be used to capture the reference position. For example, one process, discussed below in conjunction with FIG. 9B, first uses a standard camera to identify the coarse location of the reference, and then the substrate and reference are second high resolution cameras. Are repositioned as needed for image capture, ie to identify a reference micron-scale location. FIG. 2B illustrates some optional processes 229 and 231, that is, the process in which the substrate is repositioned to allow capture of multiple references. As described, the more criteria and alignment features that are used, the better the ability to detect the nonlinearity of the substrate or individual panel. In this embodiment, the substrate is controlled by a precision handler to effectively transport the reference for each panel to the camera, but in an alternative embodiment, the camera can be mounted on a motion system. , Can be transported relative to a stationary substrate to identify a reference location, or both the substrate and camera can be moved, as in the case of a split motion axis. As described by text on the left side of the figure, another embodiment uses a line scanner to image the substrate and / or its reference during substrate loading or other transport, while still other embodiments Features a camera mounted on the printhead (ie, used for very frequent “continuous” error monitoring and correction), yet another embodiment uses a non-visual detection mechanism . Regardless of the number of criteria and / or detection mechanisms, once the position details are identified and any errors are confirmed (233), the stored template is read (235) and rendered. FIG. 2B illustrates that the stored template can be in the form of a recipe (236), a bitmap (237), or some other form (238) (eg, an object, vector, or other form). . Offsets, misalignments, and / or other errors can be measured on a linear basis (239), using a least squares fit (LSF) approach (240), or using some other type of error estimation mechanism (241). ) Can be confirmed. For example, one embodiment uses two criteria per panel. The reference may be in the form of a complex shape (eg, the “cross” depicted by FIG. 9A) and used to determine panel rotation, angular offset, and X and Y scaling errors over a given panel. . Printer control data can then be adjusted or rendered from the template using linear or affine transformations. With respect to the LSF approach, non-linear errors in the panel or across the substrate can be detected and modeled by polynomials in one or more dimensions, eg, a monitoring processor operating under software control (eg, multiple Fit the curve to the detected error (using an error measure against the criterion). Adjustments are then made according to a polynomial or affine transformation to assign printer grid firing decisions and associated nozzle parameters. Equipped with a mechanism (242) for mapping the template to actual substrate locations, the supervisory processor then optionally assigns processing tasks to additional processors (eg, cores of multi-core processors). In one embodiment, a multi-core processor having 12 or more processing cores is used to divide the rendering (or conversion) task and accelerate the processing. In another embodiment, hundreds of cores are used to further accelerate the process. Such parallel processing is optional, as suggested by the use of the dashed line function box 243. Regardless of whether parallel processing is used, the system will then render the printer control data in a manner that depends on the detected substrate and / or panel position, rotation, tilt, and scale, and raster as appropriate. A scan is assigned (245). As suggested by the first optional step seen on the right side of FIG. 2B, in one embodiment (246), the latest expected droplet details for a given nozzle waveform are provided (eg, each The average and standard deviation of each unique corrugated drop volume and two-dimensional drop trajectory used for a unique nozzle). This information is used in the rendering process in a manner that is calculated to produce a precise volume filling or to ensure a homogeneous distribution of ink. As discussed below, in one embodiment, the data creates a robust measurement population and takes into account variations in the properties of the deposited liquid (eg, as a function of viscosity, ambient temperature, and other factors). As such, it is measured continuously (eg, during substrate printing). In one embodiment, as described above, the scan path is not adjusted in response to misalignment, while only the nozzle / droplet firing parameters are changed. In another embodiment, the number 247 causes rasterization to be reassessed and optimized again. For example, as nozzle firing data changes, it may be possible to perform fewer scans (or vice versa, it may be wise to add scans for optimal control over the deposited droplet volume and position). The number 247 represents that in one embodiment, the printhead offset between scans can be reassessed as needed. With the number 248, a consistency check can also be performed on printer control data that has already been partially adjusted from a spatial perspective to account for errors. For example, in one embodiment, printer control data is first rendered in dependence on detected substrate-printer variations, and later identifies fluid wells (or other discrete areas of the substrate) and A software process is invoked to “check” ink density and fill. If the conversion will deposit too much or too little ink, the software process will smooth or otherwise adjust the printer control data to eliminate errors, thereby within the desired norm. Maintain total ink fill and ink density. In yet another embodiment, the “edge” of the continuous layer (eg, encapsulated layer) is assessed 249 to maintain the desired edge profile following error adjustment. That is, due to the diffusive nature of the deposition fluid, it may be desirable to increase / decrease the ink density at the layer edge to the primary homogeneous layer thickness up to the intended layer boundary, and in such embodiments, the edge characteristics are error Can be assessed or adjusted after error processing to spatially transform the printer control data to address. Edge processing as referenced represents another type of consistency check and has been extensively discussed in the aforementioned US Provisional Application No. 62/019076 (ie see FIGS. 7A-7E). Although this provisional application discusses the use of the described techniques for encapsulating layers, these teachings are intended to enhance the accuracy of any deposited layer that is patterned or otherwise. Note that can be applied to: Many variations of these principles will occur to those skilled in the art. Finally, when processing is complete, the rendered printer control data, including any raster scan information, is sent to the printer by the numbers 251 and 253, and the method ends. As described above, in one embodiment, it is generally desirable to perform this process on a substrate in about 2 seconds or less for each substrate or panel processed by the system.

  As described above, layer information may be received, stored, and rendered in many forms. FIG. 2C provides an example of a “recipe editor” that can be used to generate and edit layer descriptions to define “ideal” printing. Such data can be stored (with or without some form of pre-processing) to generate template data that will be adjusted or transformed for each new substrate.

  More particularly, FIG. 2C shows a screenshot 261 of a software user interface (“UI”) that provides a panel recipe definition or edit. Such a UI can be used in an integrated manner with a printer or manufacturing system, at an earlier design time (eg, as part of a computer-aided design system), or in some other manner. As represented by screen shot 261, the software application advantageously mimics the layout of the board or print area and provides a mechanism for editing that layout. The UI display area 263, seen on the left side of the screenshot, encompasses three different panel sizes that are variously implemented between the six depicted panels. In this example, each of the three different panel sizes can have its own recipe. A second UI display area 265, seen in the lower right corner of the screenshot, provides a window for array definition, "Subpanel 1", "Subpanel 2", "Subpanel 3", "Subpanel 3H1V0", "Subpanel 3H0V1 ”and“ Subpanel 3H1V1 ”, including the Cartesian location of each of the depicted six panels, and a link (for each panel) to the corresponding recipe (in this case, achieved by these names) For example, “subpanel 3”) and the latter three panels (“subpanel 3H1V00”, “subpanel 3H0V1”, and “subpanel 3H1V1”) all use the same recipe (“subpanel 3”) Featuring respective horizontal and vertical offsets relative to the substrate. The top of these definitions is seen magnified (ie, because it is “selected”). A second UI display area 265 provides for graphical selection and removal of any given panel (right side) and data expansion and manipulation of any desired panel (such as panel 1 as shown). Please note that. For example, FIG. 2C shows that the top panel “Sub-Panel 1” has its upper left corner positioned 50 millimeters from the left and upper edges of the printable substrate area, respectively, so that it has a size of 825 × 500 millimeters. As well as (ie, via a selected “Corner Rounding” value) to indicate that it is defined to have selective corner rounding. One or more reference expected locations can be identified using the third UI display area 269. As described above, based on the deviation between the expected reference position (relative to the printer or print transport path) of a particular panel and any actual (ie, detected) reference position, printing (and any Printing is effectively defined relative to such criteria so that a given panel's print area) can be rendered. As suggested by the screen shot 261, different fiducials or fiducial feature points can be identified, for example, to detect scaling errors or slopes. In other embodiments, if desired, the third UI display area 269 may then be of a reference shape or board feature that matches the detected (imaged) shape that the software identifies on each board at run time. Can be configured to accept definitions. Ultimately, the UI is considered to provide a fourth UI display area 267 that allows definition of substrate size and thickness. In one embodiment, the printed layer thickness is common across all panels (eg, for each layer of each panel), but this need not be true for all embodiments. In one contemplated embodiment, the thickness can vary from panel to panel and can be affected using different local halftoning densities. For each layer, printing is performed as an integral print job across the substrate, eg, through multiple scans, and then the substrate and its wet ink are transported to the curing chamber to produce a permanent layer. The post-print curing process is also typically performed in a controlled atmosphere so as to prevent particulate matter, oxygen, or moisture contamination. Other layers can be added after the deposition liquid is cured or otherwise hardened.

As described above, for example, layer data defined using a recipe editor is suitable for embodiments, for example, as a bitmap, as raw recipe data, using a vector representation, or in some other manner. Can be stored in a number of forms. In one embodiment, the recipe data is used to create a grayscale image (eg, an array of 8-bit values for each location on the substrate), where each grayscale value in the image is a corresponding substrate location. Represents the density of the ink applied, and each grayscale value is used to generate local halftoning and build layer thickness (ie, according to the printing grid). Examples will help here. In one embodiment, a matrix of 8-bit grayscale values is defined with rows and columns of 8-bit values from “0” to “225”. Each value, for example “235”, corresponds to a unit area of the substrate and maps to a specific layer thickness. For example, the value “235” may map to a thickness of 7 microns in the finished layer (depending on process and material). If all grayscale values in the matrix share the same value, this can map to a relatively homogeneous layer, but the value varies in thickness across the panel, and sub-geometry (e.g., layer Can be varied to replenish the underlying structural voids), to adjust edge accumulation, or for any desired effect. The grayscale value assignment is “estimated” to the desired thickness (with variable local halftoning selected depending on the ink and process parameters in a manner that adapts the grayscale value to the desired thickness). Can be either pre-calibrated to the desired thickness (with halftoning having a fixed relationship with the grayscale value). In one embodiment, the correlation between gray scale value and thickness is measured after test layer formation (eg, after curing or drying) so that the deposited ink density is closely related to the final layer thickness. The Based on the measured data (or other feedback), the software can then map the droplet density to the desired grid pitch using, for example, the following equation:

In this equation, the in-scan pitch represents the interval between the drop opportunities in the first direction of relative motion between the print head and the substrate, and the cross-scan pitch is substantially perpendicular to the first direction ( Represents the interval between the chances of falling in the direction (or otherwise independent), and the parameter h (x100) is a grayscale value in percentage. In one embodiment, this relationship may vary over time, with emphasis on highlighting the most recent measurement, and therefore due to dynamic factors such as process or temperature, certain machine or ink For details, ink details can be continually remeasured to create robust experimental data for nozzle life or for other factors. Clearly, many embodiments are possible, as this example shows, to generate a “template” from recipe data, or to render a print job from a template in response to errors detected at runtime. In any of these, many different types of processing can be performed.

  3A-B and 4A-G are used to discuss various rendering processes. As described variously above, in one embodiment, errors representing linear distortion of the entire substrate or panel can be detected and corrected, while in other embodiments, the entire substrate or any panel can be corrected. Errors representing non-linear distortion can be detected and corrected.

FIG. 3A presents a flowchart 301 where a mapping between a detected panel terrain and a printed grid is first defined (303) to adjust the detected error. This mapping is generated to adapt the rendering of the stored “template” (or raw layer data) instance to the detected distortion. For example, referring briefly back to FIG. 1B, in the case of a substrate position offset, such a grid would have a pixel firing pattern relative to the actual panel position (ie, to fit the error) in a manner corresponding to the offset vector 123. Would be used to define In the example of FIG. 1C, the mapping may identify or specify a portion of the print grid that is sufficiently sized to represent a printed image that is rotated by an angle α. In the example of FIG. 1D, the mapping is large enough to accept prints that are scaled by (k ₁ ) X and (k ₂ ) Y. Generally speaking, a transformation is defined to encompass a desired printable area of a particular implementation of a product array or substrate. For example, if substrate distortion increases or decreases the substrate size relative to the template print image, the printing grid may feature more or fewer nozzle positions that represent areas that have increased, decreased, or shifted. The transform is then used to render the print data in a manner that aligns with any substrate (eg, panel) error. For example, if a pre-rendered template is stored for repeated printing runs, the template supports mapping in a manner that fits the print image to any detected run-by-run position or alignment error. Can be transformed (ie, rotated, scaled, offset, tilted, etc.) in a manner (and thus corresponding to errors). The transformation identifies (305) that portion of the printed grid corresponding to the detected panel position, and then uses the identified grid points, but is cached in a manner that is adjusted for detected errors. This is done by modifying the template data. For example, in one embodiment, as referenced by the numeral 307, if the preprocessing generated a firing decision of P (x, y) at each point on the printed grid, it matches the detected product position. Modified printer control data for printing grid nodes can be determined depending on the analysis of one or more of the position error vector and the preprocessed firing decision. This dependency state is the closest node (308) of the template data, the offset according to the error vector, or the misalignment or position error can be a non-integer multiple of the grid pitch, so two, three, Four or different numbers of launch decision weighting functions (309) can be formed. In one embodiment, the firing decision is based only on a single node of the printed grid represented by the original template (ie, an offset according to the error vector), and in the second embodiment, 4 Based on the weighted average of the two nearest nodes.

_Where P _nw (x, y), P _ne (x + 1, y), P _sw (x, y + 1), and P _se (x + 1, y + 1) are the north-west, north-east, south-west, and south-east grid points, respectively. Represents a decision, x and x ′ and y and y ′ are related according to the detected error. Clearly, this is only one error guarantee algorithm and many alternative algorithms will be readily conceived by those skilled in the art.

  For example, taking into account errors, a hypothetical grid point {P ′ (x ′, y ′)} (representing a specific grid point of the transformed panel) is a specific NE and SE grid point of the cached template. If the NE, SE, NW, and SW points are closest to each of the (printing pixels) and correspond to points that are not very close to a particular corresponding NW and SW grid point (printing pixel), respectively, ( 1, 1, 1, and 0), the printed grid point {P ′ (x ′, y ′)} is {0.40 * (1) + 0.40 * (0) + 0.10 * (1) + 0.10 * (1) = 0.60> 0.50} and is weighted according to the error distance to these four nodes (eg, launch decision of “1” is P ′ ( x ', y')) firing decisions (1, 1, 1, and As a function of) it can be assigned a binary firing decision (1 or 0). Again, it should be noted that this mathematical relationship is exemplary only and many algorithms can be used to assign nozzle firing decisions. As described, in one detail considered embodiment that uses a cached template, the firing decision processes the nozzle firing decision to ensure a consistent number of droplets in the transformed image. Based on only the nearest single overlapping print pixel from the transformed image, with a software anti-aliasing process used. In one embodiment, an affine transformation is used as described. Regardless of the correction method, when printer control data with adjusted errors is created, it is stored in memory as a new or customized print image (311) and the completed and rendered printer control data is available for execution. Is sent to the printer (313). Following layer completion (317), the system is then ready for the next layer or new product run or substrate (eg, using the same transformation).

  Note that number 310 refers to the (optional) wrapping process of the chain line used to ensure deposition consistency. For example, a particular point P ′ (x ′, y ′) will fit within the fluid well used to hold the liquid for use as the light generation layer of the display, and the weighted error function will be relative to the template image. If we seek to rely on template print pixels that fall outside that well, the weighting function can instead be shifted to rely on other print pixels associated with unfinished print wells (i.e. Wrapped). Such a wrapping process can also optionally be performed on each core or processor in the parallel processing environment, for example, borrowing the print pixels of other processors to maintain local consistency. Other embodiments are possible. For example, the number 313 is such that, in some embodiments, the template image is shifted or distorted according to the error, and then a predetermined area (eg, fluid well) of the substrate is processed and inspected for compliance within a threshold. Shows that the software daemon is called. One such process (i.e., anti-aliasing 314), following the error shift, the number of drops (or the average of each drop) to ensure that the total liquid still meets the required criteria. Including simply examining the expected total volume from the stored data representing the volume). In another variation, nozzle data, such as validation data, expected drop volume or trajectory, or other data, to ensure desired total value consistency (eg, to promote layer homogeneity) As such, adjustments to printer control data can be made as needed to provide deposition consistency. As described, many different processes can be used.

  FIG. 3B illustrates another flowchart 351 used to describe the processing of the stored template. More specifically, information representing the desired layer is first loaded from the template (353). In one embodiment, this information includes raw recipe data, in the second embodiment, the information includes a bitmap representing a nozzle firing command, and in a third embodiment, the information includes, for example, these 2 Represented by some other form that is intermediate between one form. The system then calculates (355) the transformation parameters needed to map the print represented by the template across the actual (detected) panel or substrate positions. These parameters are then applied to generate a transformed representation (357). As suggested by optional process block 359, in one embodiment, rendering of printer control data can be performed directly on recipe data (ie, stored as a template) at run time according to the conversion parameters. As described, in one embodiment, one or more affine transformations can be used (360). The transformed representation is then output as a rendered bitmap (361) and used to generate raster data (363). The transformed representation can be checked 363 for unit area ink density or total volume consistency, or for other quality control reasons. Upon receiving any correction, this data is then output to the printer (363) and the method ends (365).

  FIG. 4A provides an example 401 that is used to describe a transformation as applied to one exemplary printed grid and cached bitmap. The print grid for an ideal (ie error free) “bitmap” print image is referenced by numeral 403. Each “box” in the print grid (eg, by the number 404) represents a print “pixel” defined by the grid point, and an “X” in the box moves the printhead relative to the underlying substrate. Represents that the printing nozzles eject droplets at the corresponding lattice points. Similarly, a “box” or pixel 405 that is considered empty (ie, does not have an “X”) indicates that the print nozzle will not fire over that position. The top of FIG. 4A shows a representation of the print head 407 where individual triangles (as identified by the numeral 409) represent each nozzle. As the print head moves relative to the substrate, each given nozzle may pass through a series of print pixels (discrete drop firing points) at different times. As an example, when the nozzle 410 is moved downward relative to the page of the drawing, the area corresponding to the print pixel 404 (fired there) and later the area corresponding to the print pixel 405 (not fired there). pass. Bitmap 403 is predefined and stored in memory as a template for use in a repeatable manufacturing process.

  FIG. 4B provides an example 411 where the product case is considered to be in an offset position relative to the template, for example, due to substrate position error. Based on this offset (413), the area that should accept printing for accurate alignment is represented by numeral 412. If the nozzle firing command provided by the ideal print image 403 remains unadjusted, printing will occur in a manner that is misaligned with the underlying product geometry. Therefore, it is desirable to adjust the template so that printing occurs at the appropriate location. In this regard, the template from FIG. 4A can be distorted at run time (eg, by position shifting) to be dynamically aligned with individual substrates, panels, or other product areas to reduce errors. .

  FIG. 4C provides an example 421 of a portion of a printed grid 422 that overlaps the detected product location. Note that in this embodiment, this is the same printed grid as represented by FIG. 4A, but with a node corresponding to the detected product position. Once again, each depicted node defines a print pixel that will be passed by the printhead 407 in its motion relative to the substrate, with each “box” (such as print pixel 419) at the printhead nozzle and nozzle firing time. Correspond. In one embodiment, a particular nozzle is pre-determined (eg, a scan associated with a template will also be used for a particular product case), which is necessary for all embodiments. It is not done. However, it should be noted that the position offset (413) may not exactly match the printhead nozzle in-scan or cross-scan pitch. Thus, the software effectively overlays the detected product positions against the print grid 422 and makes a firing decision for the overlay based on the shifted version of the original template image according to the offset 413. Each print pixel in this embodiment may correspond to (overlap) a maximum of four print pixels of the original template print image, depending on the offset 413.

  For example, FIGS. 4B and 4C show a halfway point between 4/2 template print pixels (considering error 413) so that the detected product location print pixel 419 is represented by the numbers 415 and 417. Two cases corresponding to are shown. Note that for various embodiments, any number of nearby pixels may be applied in a weighting function. In one embodiment, only a single “closest” pixel is used, and in the second embodiment “two” adjacent pixels can be used (eg, by the number 417), while the first In three embodiments, a different number of nearby pixels (eg, “9”) can be used, ie, the disclosed technique is limited to consideration of a weighted average of four template print pixels. Should not be considered. The two printed pixel embodiment 417 can represent any number of pixels regardless of whether it is greater than 2, less than 2, or equal to 2, and based on a weighted average of multiple pixels. Fulfills the function. Based on the referenced pixel set, a decision is made for the print pixel 419, and once done for all grid points seen in FIG. 4C, this is the data from FIG. 4A to correspond to the actual product position. Effectively "convert". Once the process is complete, rasterization (eg, scan planning) is then performed as appropriate, and printing is then performed based on the detected product position. The above process accurately maps any ink density represented by the original template print image onto the shifted product geometry as detected in a particular print run.

  4D and 4E provide examples 431 and 441, respectively, illustrating bitmap processing in case of unintended product (array, product, substrate, or panel) rotation. As seen in FIG. 4D, numeral 435 assumes that the substrate or a portion thereof is misaligned by an angle α. Thus, for proper printing, the print nozzle firing pattern should also be distorted according to this same value, as represented by contour 433. FIG. 4E highlights the printed grid area corresponding to the detected product / panel position. Note that while the desired printing area is rotated relative to the substrate, the print head 407 and nozzles (nozzles 409, etc.) can each define a regular printing grid into which droplets can be fired. Once again, for each grid point (or print pixel) in the printed grid, the software shifts and overlays the original template image, then any of the original templates to obtain a weighted pixel firing decision for that grid point Weight superimposed print pixels. This operation is represented, for example, with reference to print pixel 443 in FIG. 4E and the associated (exemplary) processing of four template grid points 445 or two template grid points 447. Once again, the two grid point example 447 is simply cited to show that the alternative embodiment can use any number of references to the cached template print image.

  4F and 4G provide examples 451 and 461, respectively, illustrating the processing in case of scaling error. Once again, the number 403 refers to a template for a particular terrain and the number 455 represents the footprint that should accept printing to properly align the new layer with the desired product footprint according to the scaling vector 457. Represent. In such an embodiment, a new printed grid is effectively defined according to the scaling vector by a mathematical multiplication process. Note that printed grids also potentially use variable offsets according to substrate position, in addition to being effectively scaled. Assuming a two-dimensional increase in the printed product or panel footprint, the effect is as if the pixels of the overlaid template image are increased in size (eg, represented in FIG. 4G). Compare the rendered size of such template print image pixels 465 and / or 467 to the print image pixels 445 and 447 depicted in FIG. 4E).

  4H and 4I are explanatory diagrams for tilt compensation. The number 475 represents the footprint that should accept printing to properly align the new layer with the desired product footprint according to the tilt adjustment formula 477. Note that the amount of tilt varies according to the height of the object being distorted. If the error being compensated corresponds to a uniform tilt across the substrate, the amount of distortion in the “Y” dimension (left to right across the figure) is the “X” dimension value obtained from the substrate edge. Fluctuates depending on It should be noted that the X coordinate transformation in this example is uniform (ie, no change), but in general, the results will vary depending on the tilt orientation. The template is scaled according to the scaling vector by a mathematical multiplication process according to Equation 477 and effectively superimposed on the printed grid corresponding to the detected product size. The firing value for each printed grid point of the superimposed grid is calculated according to weighting criteria based on (for example) one, two, four, or another number of overlapping or neighboring printed pixels, Equation 477 (ie, Reciprocal of scaling). This is represented in FIG. 4 by the depicted match of the print grid points 483 to the weighted and positionally superimposed print pixels 485 or 487 of the template. Again, each grid point represents a possible firing position of the nozzle (eg, referenced by numeral 409) of the print head 407 as it is moved perpendicular to the page of the drawing.

  In the depicted example, there is a print pixel adjacent to the print grid footprint that is not necessarily weighted by a consistent number of template print pixels (e.g., 4) relative to other portions of the print grid. Note that there are. In order to avoid a lack of representation of the droplets adjacent to the new footprint area border region, the edge grid points are advantageously algorithms that ensure proper droplet density (eg, opposite the footprint) Pixel wrapping from). In situations where the depicted footprint (eg, 475) represents the desired layer boundary, then, if desired, the edge to ensure a well-defined edge, as previously referenced. A process (eg, fencing) can be applied. Other techniques and variations are possible.

  In one embodiment, it has been previously described that a bitmap can consist of gray scale values (eg, 8-bit or other multi-bit values), each value representing a volume of ink or a desired thickness. The process described above can also be applied to such grayscale values simply by weighting each grayscale value of the template with an appropriate distance scale and grayscale scale. For example, again, according to the normalized overlap region, 40% is mapped to each of the template NE and SE grid points (printed pixels) and 10% of the template NW and SW grid points are detected. Using the example of a product geometry hypothetical grid point {P ′ (x ′, y ′)}, the NE, SE, NW, and SW points are (235, 235, 150, and 0, respectively). ) Has a gray scale determination of {P ′ (x ′, y ′)} is {0.40 * (235) + 0.40 * (235) + 0.10 * (150) A gray scale value of 203 can be assigned according to + 0.10 * (0) = 203}. This scale can also be converted directly into a binary firing decision, if desired, by comparison to any desired threshold by software for individual printed grid annotation (eg, “1 ] = 203> Th, fire. Note that these relationships are exemplary only, and many algorithms can be used to assign nozzle firing decisions in a customized print image. As described above, in at least one considered system, droplet details per measured nozzle, per drive waveform can be taken into account in the process (eg, assuming a desired printhead scan offset). Thus, the fact that the nozzle corresponding to P ′ (x ′, y ′) fires a heavy drop, eg, volume = 12.4 pL, replaces the drop of P ′ (x ′, y ′). Can be relied upon in assigning different nozzles to fire, or conversely, for P ′ (x ′, y ′), the scan can be adapted to employ different printhead offsets and thus different nozzles. Can be re-planned). Other variations are possible.

  A method for converting a printed image is introduced, wherein the present disclosure is associated with printing and printheads used in industrial printers to produce one or more layers of a product. An exemplary circuit will be discussed.

  In the figures discussed above, a simplified printhead (eg, referenced by numeral 407 in FIGS. 4A-4I) has been depicted as having a relatively small number of nozzles, eg, twenty. In fact, especially for typical manufacturing operations for large products, the printhead may have a much larger number of nozzles, for example thousands, arranged in multiple rows. For example, a plurality of such prints mounted in a common assembly such that, in total, thousands or tens of thousands of print nozzles are used to eject material across a relatively wide strip of substrate. There can be a head. This structure provides high quality ink droplet delivery with a highly accurate position resolution. Each printhead is expected to deposit a nominal amount of ink per droplet (eg, 10 picoliters or 10.00 pL), but in practice the number is determined by nozzle position, droplet velocity, and droplet ejection. Just as the trajectory can vary, it can vary from nozzle to nozzle. These variations can potentially cause defects in the deposited layer that, if left unchecked, can be carried over to the quality of the finished product.

  In one embodiment, the droplet details of each nozzle are measured and used to create a statistical model of each nozzle's performance for each of these parameters in order to more accurately produce the desired layer. Repeated measurements effectively average measurement errors and help to develop a relatively accurate understanding of the average and sigma of each parameter, and various techniques are used to address the variations referenced above. The Generally speaking, these techniques take advantage of variation and achieve specific ink fills, ink fill densities, and drop distributions based on measured averages per nozzle or drop and associated sigma. Providing precisely planned droplet deposition that is aimed at. Halftoning (i.e., ink density) and / or template adjustment can incorporate such variations so that nozzle and / or droplet position and / or volume variations that do not work properly can be corrected (or Can be corrected). “Half toning” as used herein, ie, even if “tone” is not used in the correct sense (ie, “ink” or deposited liquid is typically colorless) Note that it refers to varying the drop density (eg, the number of drops per group of printed grid nodes) to affect the layer thickness. In one technique described below, multiple (alternative) electrical drive waveforms vary the target drop volume and position (including at least one selection approximating the ideal target drop volume and position). / Available for selection on a nozzle-by-nozzle basis to provide a selective ability to deliver. Statistical measurements can be made for each such waveform and for each nozzle to provide high planning accuracy. These measurements can be updated or re-executed over time to accommodate changes in ink properties (eg, viscosity) and to account for temperature changes, nozzle clogging or lifetime, and other factors. The following sections introduce the measurement capabilities that provide this understanding. Considering briefly the above discussion regarding the use of print grids, each grid point will be associated with a nozzle of the printhead. The measurement capabilities referred to above are, for example, so that nozzles that do not operate normally are not assigned firing decisions in the printhead (and / or so that any required droplets are redistributed to other nozzles). Can be applied in this process. In addition, the difference in the “X” axis position of the droplets can be achieved through the use of referenced alternative nozzle drive waveforms to alter the droplet timing to better position the droplets or change their volume. Alternatively, it can be corrected through adjusting or amplifying the preselected waveform. Substantial additional details about these practices are provided in previously referenced applications (incorporated herein by reference).

  5A-5D are used to introduce control over nozzle firing and drive waveform selection.

  Typically, the effect of different drive waveforms and resulting droplet volume is measured in advance. In one embodiment, for each nozzle, then up to 16 different drive waveforms can be used for selective use later in providing discrete variation as selected by the software, 1k per nozzle. It is stored in a static random access memory (SRAM). With different drive waveforms at hand, each nozzle is then instructed for each droplet as to which waveform to apply through programming of data to achieve a particular drive waveform. This configuration information is stored separately by the planter (or control processor) from launch decisions that will be applied to the substrate.

  FIG. 5A illustrates one such embodiment, generally designated by the numeral 501. In particular, processor 503 is used to receive data defining the intended fill per target area for a particular layer of material to be printed. As represented by the numeral 505, the data can be a layout file or a bitmap file that defines the drop volume per grid point or position address. A series of piezoelectric transducers 507, 508, and 509 generate associated ejected drop volumes 511, 512, and 513 that depend on many factors, including nozzle drive waveforms and printhead-to-printhead manufacturing variations, respectively. During the calibration operation, each set of variables is tested for its effect on drop volume, including inter-nozzle variation and the use of different drive waveforms, taking into account the particular ink that will be used. If desired, the calibration operation can be made dynamic to respond to changes in temperature, nozzle clogging, nozzle life, or other parameters, for example. This calibration is represented by a droplet measurement device 515 that provides measurement data to the processor 503 for use in managing the print plan and the next print. In one embodiment, this measurement data is literally a few minutes, eg, for thousands of nozzles (eg, for thousands of printhead nozzles and potentially many possible nozzle firing waveforms) as an offline process. Calculated during operation, which takes only 30 minutes, preferably less time. In another embodiment, such measurements are repeated, ie, for each nozzle at different times (eg, between successive substrates in an assembly line process as the substrate is loaded and unloaded). It can be done in a manner that updates different subsets. Non-imaging (eg, interference) techniques can be used at will, potentially covering tens to hundreds of nozzles per second, potentially resulting in tens of droplet measurements per nozzle. This data and any associated statistical models (and averages) are stored in memory 517 for use in processing layout data or printed image (eg, bitmap data) 505 when received. Can do. In one implementation, the processor 503 is part of a computer that is remote from the actual printer, while in a second implementation, the processor 503 is a production mechanism for a product (eg, a system for producing a display). ) Or a printer.

To perform droplet firing for the depicted embodiment, one or more sets of timing or synchronization signals 519 are received for use as a reference, which are specified nozzles (respectively 527 528, and 529) are passed through the clock tree 521 for delivery to each nozzle driver 523, 524, and 525 to generate a drive waveform. Each nozzle driver has one or more registers 531, 532, and 533 that receive multi-bit programming data and timing information from the processor 503, respectively. Each nozzle driver and its associated register receives one or more dedicated write enable signals (we _n ) for the purpose of programming registers 531, 532, and 533, respectively. In one embodiment, each of the registers selects between a significant amount of memory, including a 1k static RAM (SRAM) that stores a plurality of predetermined waveforms, and controls the waveform generation differently. Programmable registers. Data and timing information from the processor is depicted as multi-bit information, but this information can be provided via either a serial or parallel bit connection to each nozzle (discussed below). As seen in FIG. 5B, in one embodiment, the connection is in series as opposed to the parallel signal representation seen in FIG. 5A).

  For a given deposition, printhead, or ink, the processor selects a set of 16 drive waveforms for each nozzle that can be selectively applied to generate droplets. Note that the number is arbitrary; for example, in one design, four waveforms may be used, while in another design, 4000 waveforms may be used. These waveforms may advantageously be, for example, a substantially ideal drop volume (eg, an average liquid of 10.00 pL) to provide the desired variation in output drop volume and / or position for each nozzle. Selected to cause each nozzle to make at least one waveform selection that produces (drop volume) and to provide a series of intentional volume variations from each nozzle. In various embodiments, the same set of 16 drive waveforms is used for all of the nozzles, but in the depicted embodiment, 16 possibly unique, each waveform provides a respective drop volume characteristic. Are pre-defined for each nozzle separately.

During printing, in order to control the deposition of each droplet, the data for selecting one of the predefined waveforms is then stored on a per-nozzle basis, each nozzle's respective register 531, 532, or 533 is programmed. For example, considering a target drop volume of 10.00 pL, the nozzle driver 523, through writing data to register 531, out of 16 waveforms corresponding to one of 16 different drop volumes. Can be configured to set one of Once the drop volume and associated distribution per nozzle (and per waveform) is registered by the processor 503 and stored in memory to help produce the desired target fill, the volume produced by each nozzle is Will be measured by the droplet measurement device 515. This same process can be performed for droplet positions or trajectories. The processor can define whether the particular nozzle driver 523 wants to output one of the 16 waveforms selected by the processor by programming register 531. The processor also uses a per-nozzle delay or offset to firing the nozzles for a given scan line (eg, speed or trajectory error to align each nozzle with the grid traversed by the printhead The registers can also be programmed to correct for errors including, and for other purposes, and this offset is the use of a specific nozzle (or firing waveform) by a programmable number of timing pulses for each scan Is achieved by a counter. In providing an example, if the result of a drop measurement indicates that one particular drop tends to have a lower velocity than expected, the corresponding nozzle waveform is triggered earlier (eg, Can be advanced in time by reducing the dead time before the active signal level used for piezoelectric actuation, and conversely, the result of a droplet measurement is If it indicates having a high speed, the waveform can be triggered later, and so on. Other examples are clearly possible, for example, in some embodiments, by increasing the drive strength (ie, the signal level and associated voltage used to drive the piezoelectric actuator of a given nozzle). Slow drop velocity can be hindered. In one embodiment, the synchronization signal delivered to all nozzles occurs at a defined time interval (eg, 1 microsecond) for synchronization purposes, and in another embodiment, the synchronization signal is, for example, a print Adjusted for printer motion and substrate topography to fire a progressive relative motion of 1 micron between the head and the substrate. A high-speed clock (φ _hs ) is operated thousands of times faster than the synchronization signal, eg, 100 megahertz, 33 megahertz, etc., and in one embodiment a plurality of different clocks or other timing signals (eg, strobe signals) Can be used in combination. The processor also programs values that define the grid spacing, and in some implementations the grid spacing is common to the entire set of available nozzles, but this need not be true for each implementation. For example, in some cases, a regular printed grid can be defined where all nozzles fire "every 5 microns". The grid can be unique to the printing system, the substrate, or both. Thus, in one optional embodiment, the identification is performed using a synchronization frequency or nozzle firing pattern that is used to effectively transform the printed grid to match the substrate terrain that is a priori unknown. A printing grid can be defined for a number of printers. In another contemplated embodiment, the processor may then select (as required) a new grid spacing that is read from all nozzles (eg, to define an irregular printed grid). In addition, memory is shared across all nozzles that allows the processor to pre-store several different grid spacings (eg, 16) that are shared across all nozzles. For example, in implementations where the nozzle fires for all color component wells of an OLED (eg, to deposit a non-color specific layer), three or more different grid spacings may be generated by the processor Can be applied continuously in a round-robin fashion. Clearly, many design alternatives are possible. The processor 503 can also dynamically reprogram each nozzle's register during operation, i.e., a sync pulse as a trigger to trigger any programmed waveform pulse set in that register. Note that if new data is received asynchronously before the next sync pulse, the new data will be applied with the next sync pulse. The processor 503 also controls the start and speed (535) of the scan in addition to setting parameters for the sync pulse generation (536). In addition, the processor controls the optional rotation (537) of the printhead. In this way, each nozzle can be used together at any time (ie, with any “next” sync pulse) using any one of 16 different waveforms for each nozzle (or (Simultaneously) and the selected firing waveform can be dynamically switched between firings during a single scan with the other of any of the 16 different waveforms.

FIG. 5B shows additional details of the circuit (541) used in such an embodiment to generate an output nozzle drive waveform for each nozzle, the output waveform being “nzzzl-drv. expressed as “wvfm”. More specifically, circuit 541 receives an input of a synchronization signal, a single bit line that carries serial data (“data”), a dedicated write enable signal (we), and a high-speed clock (φ _hs ). Register file 543 provides data for at least three registers, each carrying an initial offset, a grid definition value, and a drive waveform ID. The initial offset is a programmable value that adjusts each nozzle to align with the start of the print grid, as described. Considering implementation variables such as multiple printheads, multiple rows of nozzles, different printhead rotations, nozzle firing speeds and patterns, and other factors, the initial offset should take into account delays and other factors , Each nozzle droplet pattern can be used to align with the start of the printing grid. The offset can be applied in different ways across multiple nozzles, for example, to rotate the printed grid or halftone pattern relative to the substrate terrain, or to correct substrate misalignment. Similarly, as described, offsets can also be used to correct abnormal speed or other effects. The grid definition value is a number that represents the number of sync pulses that are “counted” before the programmed waveform is triggered, and for implementations that print flat panel displays (eg, OLED panels) The region has one or more regular spacings for different printhead nozzles, possibly corresponding to regular (constant spacing) or irregular (multiple spacing) printing grids. As mentioned above, in some implementations, the processor has its own 16-entry SRAM to define up to 16 different print grid spacings that can be read into the register circuit for all nozzles on demand. keep. Thus, if the print grid spacing value is set to 2 (eg, every 2 microns), each nozzle will be fired at this spacing. The drive waveform ID is a selection value that is used to select one of the pre-stored drive waveforms for each nozzle. In one embodiment, the drive waveform ID is a 4-bit selection value and each nozzle is uniquely dedicated to store up to 16 predetermined nozzle drive waveforms, stored as 16 × 16 × 4B entries. It has a 1k byte SRAM. Briefly, each waveform consists of 16 discrete signal levels, and each of the 16 entries for each waveform contains 4 bytes representing a programmable signal level, these 4 bytes being the high-speed clock. Represents a 2-byte resolution voltage level and a 2-byte programmable duration used to count the number of pulses. Thus, each programmable waveform can be programmed from (0-1) discrete pulses up to 16 each of programmable voltage and duration (eg, duration equal to 1-255 pulses of a 33 megahertz clock). It can consist of up to discrete pulses.

The numbers 545, 546, and 547 specify one embodiment of the circuit that shows how a prescribed waveform can be generated for a given nozzle. The first counter 545 receives a sync pulse to initiate an initial offset countdown triggered by the start of a new line scan. The first counter 545 counts down in micron increments, and when it reaches zero, a trigger signal is output from the first counter 545 to the second counter 546. This trigger signal essentially initiates the firing process of each nozzle for each scan line. The second counter 546 then implements a programmable grid spacing in micron increments. The first counter 545 is reset in conjunction with the new scan line, while the second counter 546 is reset using the next edge of the high speed clock following its output trigger. The second counter 546 activates a waveform circuit generator 547 that, when triggered, generates a selected drive waveform shape for a particular nozzle. As represented by the dashed box 548-550 seen under the generator circuit, this latter circuit is timed according to a high speed clock (φ _hs ), a high speed digital to analog converter 548, a counter 549, and high voltage amplifier 550. When a trigger from the second counter 546 is received, the waveform generator circuit retrieves the number pair (signal level and duration) represented by the drive waveform ID value and determines a predetermined analog output voltage according to the signal level value. Counter 549 is effective to hold the DAC output for the duration according to the counter. The associated output voltage level is then applied to the high voltage amplifier 550 and output as a nozzle drive waveform. The next number of pairs is then latched from register 543, etc. to define the next signal level value / duration, etc.

  The depicted circuit provides an effective means of defining any desired waveform according to the data provided by the processor 503. If it is necessary to conform to the geometry of the grid, or to mitigate nozzles with unusual speeds or flight angles, any particular signal level (e.g. a first “0” defining an offset to synchronization) The duration and / or voltage level associated with the signal level can be adjusted. As described, in one embodiment, the processor pre-determines a set of waveforms (eg, 16 possible waveforms per nozzle) and then defines each of these selected waveforms for each By writing to the SRAM for the nozzle driver circuit and then writing the 4-bit drive waveform ID to each nozzle register, a default selection of programmable waveforms to be applied in response to the firing decision is achieved.

  The use of multiple signal levels to shape the pulse is discussed with reference to FIG. 5C.

  That is, in one embodiment, the waveform can be predefined as a series of discrete signal levels, eg, defined by digital data, and the drive waveform is generated by a digital-to-analog converter (DAC). . Number 551 in FIG. 5C refers to waveform 553 having discrete signal levels 555, 557, 559, 561, 563, 565, and 567. As described for this embodiment, each nozzle driver includes circuitry for receiving and storing up to 16 different signal waveforms, each waveform being represented as a series of multi-bit voltages and durations, respectively. It is defined as a maximum of 16 signal levels. That is, the pulse width can be effectively varied by defining different durations for one or more signal levels, and the drive voltage is selected to provide subtle droplet size variation For example, droplet volume is measured to provide a specific volume gradual increment, such as in units of 0.10 pL. Thus, in such embodiments, corrugation provides the ability to adjust the droplet volume to approach the target droplet volume value. When combined with other specific droplet volumes and locations, such as using the techniques exemplified above, these techniques facilitate precise fill volumes per target area. In another embodiment, a predetermined waveform can be applied, and optional further waveform shaping or timing is applied as appropriate to adjust droplet volume, velocity, and / or trajectory. In yet another embodiment, the use of nozzle drive waveform alternatives provides a mechanism for volume planning so that no further waveform shaping is required.

  FIG. 5D shows yet another design 571. For example, the CPU 573 of the master computer sends print data (adjusted for error) to a print module having a plurality of print heads (eg, one of six different print heads, each having hundreds of print nozzles). Send. In the printing module, an Ethernet connection 575 receives data from the CPU and provides the data to a field programmable gate array (“FPGA”) 577. A customized “soft processor” (578) in the FPGA processes the data as appropriate and writes the data to memory (ie, dynamic random access memory or “DRAM”) 579. Unlike the embodiments described above, in the depicted embodiments, the particular form in which the waveform is written to memory and data is written or stored for use by one or more nozzles. May be an implementation decision. For example, in one embodiment, the soft processor 578 writes a specific waveform as a series of drive levels for each nozzle (see, eg, the discussion above in connection with FIG. 5C), and then the other FPGA logic 580 This data is read at a convenient time, provided to the amplifier 581 and finally provided to the associated print head 583 and the associated nozzle driver. In one implementation, the depicted FPGA 577 and DRAM 579 are components of each printhead (eg, data is transmitted separately to each printhead), but this is not the case for all embodiments. One function of logic 580 and amplifier 581 is to put the appropriate waveforms in parallel form so that they can be read together for each nozzle in parallel at the appropriate time (if necessary or as appropriate to the design). Note that it is to do. In one implementation, DRAM 579 can optionally be organized to have a separate memory for each nozzle. In this embodiment, instead of using a trigger, each nozzle receives one of 16 waveforms (i.e., one waveform is a flat indicating that a particular nozzle is not fired). Note that it is a waveform or zero drive signal). Specific drive waveforms can be dynamically supplied or pre-written by CPU 573 (ie, but can be dynamically changeable), and the run-time 4-bit value can be used to determine nozzle firing decisions and associated waveform selection. (As well as triggering the output of that waveform from DRAM 579 or logic 580). Other alternatives are possible.

  Thus, while nozzle control circuitry has been introduced as can be used in exemplary manufacturing devices, additional details are now presented regarding one possible implementation of such devices. As previously suggested, one contemplated implementation of the techniques described herein is to manufacture flat panel devices in an array, which are then cut from a common substrate. The In the discussion below, more particularly, the manufacture of solar panels and / or display devices that can be used in electronic equipment (eg, as a smartphone, smartwatch, tablet, computer, television, monitor, or other form of display). An exemplary system for performing such printing as applied to will be described. The manufacturing techniques provided by the present disclosure are not limited by this particular application and can be applied, for example, to any 3D printing application and a wide range of other forms of products.

  FIG. 6A represents several different implementation layers, collectively designated by reference numeral 601. Each one of these layers represents a possible discrete implementation of the techniques introduced herein. First, the techniques introduced in this disclosure may be implemented in the form of instructions stored on a non-transitory machine-readable medium (eg, execution to control a computer or printer) as represented by graphic 603. Possible instructions or software). Second, with the computer icon 605, these techniques are also optionally implemented as part of a computer or network, for example, within a company that designs or manufactures components for use in sales or other products. Can also be done. Third, as illustrated using the storage media graphic 607, the previously introduced techniques are different ink volumes to reduce alignment errors when acted upon, for example, by the discussion above. Alternatively, it can take the form of stored printer control instructions as data that would cause the printer to produce one or more layers of components that depend on the use of location. Note that the printer instructions can be transmitted directly to the printer, for example via a LAN. In this regard, storage media graphics can represent portable media such as, but not limited to, a RAM or flash drive that is internal to or accessible to a computer or printer. Fourth, as represented by production device icon 609, the techniques introduced above may be implemented as part of a production device or machine, or in the form of a printer within such device or machine. it can. Note that the particular depiction of the fabrication device 609 represents one exemplary printer device that will be discussed in connection with FIG. 6B below. The techniques introduced above can also be embodied as an assembly of manufactured components. For example, in FIG. 6A, several such components are depicted in the form of an array 611 of semi-finished flat panel devices that will be sold separately for incorporation into the final consumer product. The depicted device may have, for example, one or more photogenerating or encapsulation layers, or other layers that are fabricated depending on the techniques introduced above. The techniques introduced above are also for example in the form of a display for a portable digital device 613 (eg, an electronic pad or smartphone), a television display screen 615 (eg, OLED TV), a solar panel 617, or other As a type of device, it can also be embodied in the form of an end consumer product as referenced.

  FIG. 6B shows one contemplated multi-chamber fabrication apparatus 621 that can be used to apply the techniques disclosed herein. Generally speaking, the depicted apparatus 621 includes several general modules or subsystems, including a transfer module 623, a printing module 625, and a processing module 627. Each module can be printed by the printing module 625, for example, in a first controlled atmosphere, and other processing, such as inorganic encapsulating layer deposition or (for example, for printed materials) ) Maintain a controlled environment so that another deposition process, such as a curing process, can be performed in the second controlled atmosphere. Apparatus 621 uses one or more mechanical handlers to move the substrate between modules without exposing the substrate to an uncontrolled atmosphere. Within any given module, it is possible to use other substrate handling systems and / or specific devices and control systems that are adapted to the processing performed on that module.

  Various embodiments of the transfer module 623 include an input load lock 629 (ie, a chamber that provides buffering between different environments while maintaining a controlled atmosphere) and a transfer chamber 631 (also a handler for transporting substrates). And an atmospheric buffer chamber 633. Within the print module 625, other substrate handling mechanisms such as a floating table for stable support of the substrate during the printing process can be used. In addition, an xyz motion system, such as a split axis or gantry motion system, can be used for precise positioning of at least one printhead relative to the substrate and a y-axis for transport of the substrate through the print module 625. Provide a transportation system. Also, for example, printing multiple inks for printing using, for example, each printhead assembly so that two different types of deposition processes can be performed in a printing module in a controlled atmosphere. It can also be used in a chamber. The printing module 625 introduces an inert atmosphere (eg, nitrogen) and is otherwise a means for controlling the presence of environmental conditioning atmospheres (eg, temperature and pressure), gas components, and particulate matter. In addition, a gas enclosure 635 for housing the inkjet printing system can be provided.

  Various embodiments of the processing module 627 can include, for example, a transfer chamber 636, which also has a handler for transporting the substrate. In addition, the processing module can also include an output load lock 637, a nitrogen stack buffer 639, and a cure chamber 641. In some applications, the curing chamber can be used to cure, bake, or dry the monomer film into a uniform polymer film. For example, two processes that are considered in detail include a heating process and an ultraviolet radiation curing process.

  In one application, the device 621 is adapted for mass production of liquid crystal display screens or OLED display screens, for example, the fabrication of an array of eight screens at a time (for example) on a single large substrate. These screens can be used on televisions and as display screens for other forms of electronic devices. In the second application, the apparatus can also be used for mass production of solar panels in the same manner.

  The printing module 625 can advantageously be used in such applications to deposit organic photogenerating or encapsulating layers that help protect sensitive elements of OLED display devices. . For example, the depicted apparatus 621 can be loaded with a substrate to move the substrate back and forth between the various chambers in a manner that is not interrupted by exposure to an uncontrolled atmosphere during the encapsulation process. Can be controlled. The substrate can be loaded via input load lock 629. A handler located in the transfer module 623 can move the substrate from the input loadlock 629 to the printing module 625 and can move the substrate to the processing module 627 for curing following completion of the printing process. . By repeated deposition of subsequent layers, each of a controlled thickness, the total encapsulation can be accumulated to suit any desired application. Once again, it should be noted that the techniques described above are not limited to the encapsulation process, and many different types of tools can be used. For example, the configuration of the device 621 can be varied to place the various modules 623, 625, and 627 in different juxtapositions, and additional, fewer, or different modules can be used. it can.

  Although FIG. 6B provides an example of a connected chamber or set of fabrication components, there are clearly many other possibilities. The techniques introduced above can be used to control the fabrication process performed with the device depicted in FIG. 6B, or indeed by any other type of deposition equipment.

  FIG. 6C provides a top view of the substrate and printer as may appear during the deposition process. The printing chamber is generally designated by the reference numeral 651, the substrate to be printed is generally designated by the numeral 653, and the support table used to transport the substrate is generally designated by the numeral 655. Generally speaking, it includes an x and y dimension movement of the substrate by a support table (eg, using floating support as represented by numeral 657) and generally as represented by arrow 663, traveler 661. Arbitrary xy coordinates of the substrate are reached by a combination of movements using a “slow axis” x-dimensional movement of one or more printheads 659 along the axis. As described, the floating table and substrate handling infrastructure moves the substrate and advantageously provides tilt rejection control along one or more “fast axes” as needed. Used for. Each printhead (e.g., so as to achieve column printing corresponding to printer grid points when the printhead is moved along the "slow axis" from left to right and vice versa) It is considered to have a plurality of nozzles 665 controlled separately by a firing pattern derived from the template. When relative motion between one or more printheads and the substrate is provided in the direction of the fast axis (ie, the y-axis), printing typically occurs at the printer grid points of the individual rows. Represents a strip that follows a line. The printhead can also advantageously be adjusted to vary the effective nozzle spacing (eg, by the rotation of one or more printheads by the numeral 667). It should be noted that a plurality of such printheads can be used together, oriented with x-, y-, and / or z-dimensional offsets relative to each other as desired (axis legend in FIG. 6C). 669). The printing operation continues as desired until the entire target area (and any border area) is printed with ink. Subsequent to deposition of the required amount of ink, either by evaporating the solvent to dry the ink (eg, using a thermal process) or by using a curing process such as an ultraviolet curing process Is completed.

  FIG. 6D provides a block diagram illustrating various subsystems of one apparatus (671) that can be used to fabricate a device having one or more layers as defined herein. To do. Coordination across the various subsystems is provided by a set of processors 673 that operate under instructions provided by software (not shown in FIG. 6D). As described above, to perform on-the-fly conversion to correct for board or panel errors, in one embodiment, these processors include a supervisory processor or general purpose CPU and a set of additional processors used for parallel processing. Including. In one implementation that is considered in detail, these additional processors take the form of a multi-core processor or graphics processing unit (with a GPU, eg, with hundreds or more cores). Each core is assigned a portion of the print image to transform it to distort the print image to match the substrate error. If the error is corrected independently on a panel or other product-by-product basis, each core is advantageously (ie, a particular core is a single in all assigned image data). A part or part of a single product or panel can be assigned (to perform only a conversion operation). Parallel processing and / or this form of delegation is not required for all embodiments. During the fabrication process, the processor sends data to the printhead 675 to cause the printhead to eject various volumes of ink in response to firing instructions provided by the halftone print image. The printhead 675 typically has a plurality of inkjet nozzles arranged in rows or arrays and associated reservoirs that allow ejection of ink in response to activation of a piezoelectric or other transducer. Such transducers cause each nozzle to eject a controlled amount of ink in an amount controlled by an electron firing waveform signal applied to the corresponding piezoelectric transducer. Other firing mechanisms can also be used. The printhead applies ink to the substrate 677 at various xy positions corresponding to the grid coordinates, as represented by the halftone print image. Position variation is achieved by both the printhead motion system 679 and the substrate handling system 681 (eg, causing printing to show one or more bands across the substrate). In one embodiment, the printhead motion system 679 moves the printhead back and forth along the traveler, while the substrate handling system provides stable substrate support and substrate "for example, for alignment and tilt exclusion. Provide both x "and" y "dimensional transport (and rotation). During printing, the substrate handling system provides relatively fast transport within one dimension (eg, the “y” dimension with respect to FIG. 6C), while the printhead motion system 679 is, for example, a printhead offset. Thus, it provides relatively slow transport within another dimension (eg, the “x” dimension for FIG. 6C). In another embodiment, multiple printheads can be used with primary transport handled by the substrate handling system 681. The image capture device 683 can be used to locate any reference and assist in the previously described alignment and / or error detection functions.

  The apparatus also includes an ink delivery system 685 and a printhead maintenance system 687 to assist in the printing operation. The printhead can be periodically calibrated or subjected to a maintenance process. For this purpose, during the maintenance sequence, the printhead maintenance system 687 is used to perform proper preparation, ink or gas purging, testing and calibration, and other operations as appropriate for the particular process. The Such a process is also described, for example, as discussed in a previously referenced applicant's PCT patent application (PCT / US14 / 35193) and as referenced by the numbers 691 and 692, Individual measurements of parameters such as velocity and trajectory can also be included.

  As previously introduced, the printing process can be performed in a controlled environment, that is, in a manner that presents a reduced risk of contaminants that can degrade the effectiveness of the deposited layer. To this effect, the apparatus includes a chamber control subsystem 689 that controls the atmosphere within the chamber, as represented by function block 690. An optional process variant, as described, is another inert gas atmosphere (or having a specifically selected gas and / or controlled to exclude unwanted particulate matter). Carrying out the ejection of the deposited material in the presence of the environment). Finally, as represented by numeral 693, the apparatus is also used to store halftone pattern information or halftone pattern generation software, template print image data, and other data as required. A memory subsystem is also included. For example, the memory subsystem operates for conversion of a pre-generated print image in accordance with the techniques introduced above to internally generate printer control instructions that govern the firing (and timing) of each drop. Can be used as memory. If some or all of such rendering is performed elsewhere and the task of the apparatus is to create a device layer according to the received printer instructions, the received instructions may include the printing process and / or Or it can be stored in the memory subsystem 693 for appropriate use during maintenance. As represented by numeral 694, in one optional embodiment, individual droplet details are varied through variations in the firing waveform of any given nozzle (eg, to correct nozzle anomalies). Can do. In one embodiment, a set of alternative firing waveforms is pre-selected and made available to each nozzle on a shared or dedicated basis, and optionally used in conjunction with substrate variation (error) processing 695 as described above. Can. As described, some embodiments use a predetermined scan path with compensation for errors achieved using different nozzles and / or certain nozzle drive waveforms (ie, despite errors). However, in another embodiment, print optimization (696) is performed to reassess scan path details and potentially improve deposition time.

  FIG. 7 provides another flow diagram 701 associated with some of the processes discussed. As in the previous embodiment, the number 703 first receives data representing the desired layer layout. This data defines the boundaries of the deposited layers and provides sufficient information to define the thickness throughout the layer of interest (eg, for a given panel). This data can be generated on the same machine or device where process 701 is performed, or can be generated by a different machine. In one embodiment, the received data is defined according to an xy coordinate system, and the provided information is, for example, consistent with a previously introduced x micron x y micron x z micron example of a layer. It is sufficient to calculate the desired layer thickness at any represented xy coordinate point that defines a single height or thickness applied throughout. By numeral 705, the data can be converted to a grayscale value for each print cell (or each print pixel) in a deposition region that can accept the layer. If the print cell area does not inherently correspond to an xy coordinate system that matches the layout data, the layout data is obtained to obtain a grayscale value for each print pixel (eg, average thickness data for multiple coordinate points). And / or using interpolation). This transformation may be based on desired mapping information generated using, for example, relationships or equations. The number 707 allows the gray scale value to be adjusted arbitrarily to produce a homogeneous layer (or for other desired effects). In providing an example, if it is desired to compensate for the various heights of the microstructures that may be located below the desired layer, an optional technique adds an offset to the top surface of the deposited layer. Select a grayscale value that “strengthens” the layer of interest at a particular location to effectively flatten. In one embodiment, correcting for nozzle firing anomalies (eg, in the in-scan direction) can result in more ink (eg, if a particular nozzle or set of nozzles produces insufficient ink volume) or (eg, Gray scale value manipulation can also be used to deposit less ink (if a particular nozzle or set of nozzles produces excessive ink volume). Such an optional process can assume data determined by the calibration process and / or experiment by function block 714. The number 709 then converts the grayscale value into a droplet density pattern (eg, a specific halftone pattern), and then the number 710 generates a bitmap. Error diffusion as shown by the figure can be relied upon to help promote layer homogeneity.

  FIG. 7 optionally illustrates a number of nozzle / waveform error correction processes 713 applied to ensure deposition layer uniformity and accuracy and to help adjust detected panel or substrate position errors. Indicates use of a set. To ensure accurate alignment with any sub-product layer, to create a suitable encapsulation to create a water / oxygen barrier, or to provide a high quality light generating or light guiding element for a display panel, Or, for other purposes or effects, such uniformity can be important to device quality. As stated above, the data determined (estimated) by the calibration process or experiment is the number 714 to correct for grayscale value errors, or nozzle or nozzle waveform details, or substrate or panel position variations. Can be used. Alternatively, individual nozzle drive waveforms can be planned or adjusted to correct for errors, as represented by numeral 715. In another embodiment, as described above, the nozzles can be verified or qualified (719), each nozzle being determined to meet a minimum drop production threshold or not being used. One of them is considered eligible. If a particular nozzle is considered ineligible but is tentatively selected for use, a different nozzle (or acceptable nozzle) will be identified by the numeral 716 to provide proper droplet deposition. Can be used to deposit droplets that would otherwise have been printed by nozzles deemed ineligible. For example, in one embodiment, the printhead may be configured so that if one nozzle is abnormal, different redundant nozzles can be used to deposit the desired droplets at a particular grid point. And nozzles arranged in both rows. Optionally and also, for example, using different nozzles (with a printhead adjusted in position to allow this) or increasing or decreasing the number of scans, the desired droplets are deposited. Such problems can also be used to offset the printhead and adjust the scan path in such a way that it is possible. This is represented by the numeral 717 in FIG. Many such alternatives are possible. As represented by the numbers 720 and 721, in one embodiment, a droplet measurement device (720) in which each nozzle repeatedly measures droplet parameters (so as to produce a distribution per measurement nozzle or per drive waveform). And then the software understands nozzle position errors and / or volume, velocity, and trajectory nozzle droplet averages, and the variance per expected nozzle for each of these parameters And build a statistical model (721) for each nozzle. This data, as described, to consider / verify a particular nozzle (and / or droplet) or to select a nozzle that would be used to generate each individual droplet And, alternatively, can be used to adjust or customize the template print image of each new substrate or other product array. Each such measurement / error correction process, ie, print image adjustment / customization, so that printer control data is generated and / or updated to optimize the printing process while ensuring the desired layer properties. Can be taken into account in the print plan (722), including any scan path plan and / or print grid calculation. Finally, the number 725 then generates the final print data for transmission to the printer at run time.

  8A-8D are generally used to introduce techniques for droplet measurement and verification per nozzle.

  More specifically, FIG. 8A is represented by an optical system 801 and a relatively large printhead assembly 803 (ie, a nozzle plate for each printhead 805A / 805B, each with a number of individual nozzles, eg, 807. Provide an illustrative figure depicting In a typical implementation, there are hundreds to thousands of nozzles. An ink supply (not shown) is fluidly connected to each nozzle (e.g., nozzle 807), and a piezoelectric transducer (also not shown) is an ink liquid under the control of an electronic control signal per nozzle. Used to squirt drops. The nozzle design maintains a slight negative pressure of ink at each nozzle (e.g., nozzle 807) so as to avoid flooding the nozzle plate, and the electronic signal for a given nozzle activates the corresponding piezoelectric transducer, Pressurize the ink for a given nozzle, thereby ejecting one or more droplets from the given nozzle. In one embodiment, the control signal for each nozzle is typically 0 volts, and a positive pulse or signal level at a given voltage causes a particular nozzle to emit droplets (one per pulse) for that nozzle. Used for. In another embodiment, different tuned pulses (or other more complex waveforms) can be used for each nozzle. However, in connection with the embodiment provided by FIG. 8A, the drop is collected from the printhead downwardly by the discharge crucible 809 (ie, representing the z-axis height relative to the three-dimensional coordinate system 808 “ It should be assumed that it is desired to measure the droplet volume produced by a particular nozzle (eg, nozzle 807) emitted (in the direction of h). In a typical application, the “h” dimension is typically about 1 millimeter or less, with thousands of nozzles having each drop individually measured in this manner in an operating printer. Note that there are (for example, 10,000 nozzles). Thus, precisely each drop (ie, a drop originating from a particular one of thousands of nozzles in a large printhead assembly environment within an approximately 1 millimeter measurement window as described). Certain techniques are used in the disclosed embodiments to precisely position the optical assembly 801, the printhead assembly 803, or both elements relative to each other for optical measurements.

  In one embodiment, these techniques may be used to (a) precisely position the measurement region 815 immediately adjacent to any nozzle that is to generate a droplet for optical calibration / measurement (eg, a dimension plane). Utilize a combination of xy motion control (811A) of at least a portion of the optical system (within 813) and (b) sub-surface optical recovery (811B) (eg, thereby, despite a large printhead surface area) , Allowing easy placement of the measurement area next to any nozzle). Thus, in an exemplary embodiment having about 10,000 or more print nozzles, the motion system is as discrete as (for example) 10,000, close to the discharge path of each nozzle of the printhead assembly. It is possible to position at least a portion of the optical system in position. In one embodiment, a continuous motion system or system with even finer positioning capabilities can be used. As discussed below, two considered optical measurement techniques include shadowgraphy and interferometry. With each, an optical so that a precise focus is maintained on the measurement area, so as to capture the droplets in flight (for example, in the case of shadowgraphy, effectively image the shadow of the droplets) The part is typically adjusted in place. Because typical droplets can be about a few microns in diameter, the optical arrangement is typically very precise and presents challenges with respect to the relative positioning of the printhead assembly and measurement optics / measurement area Please note that. In some embodiments, to assist in this positioning, the optical section can be placed close to the measurement area without interfering with the relative positioning of the optical system and the printhead. (Mirrors, prisms, etc.) are used to orient the light capture path for sensing below the dimension plane 813 arising from the measurement region 815. This allows effective position control in a manner that is not limited by the millimeter order deposition height h at which droplets are imaged inside or the large x and y widths occupied by the monitored printhead. . With interferometry-based drop measurement techniques, separate rays incident on a small drop from different angles create an interference pattern that can be detected from a viewpoint substantially perpendicular to the optical path, and thus such a system The inner optics capture light in a manner that utilizes sub-surface optical recovery to measure droplet parameters, but from an angle that is approximately 90 degrees off the source beam path. Other optical measurement techniques can also be used. In yet another variation of these systems, the motion system 811A optionally and advantageously selectively engages and disengages the drop measurement system without moving the printhead assembly during drop measurement. Made to be an xyz motion system. In brief, for industrial fabrication devices with one or more large printhead assemblies, each printhead assembly has one or more maintenance functions to maximize manufacturing uptime. Considering that it will sometimes be “stationed” at the service station, and considering the pure size of the printhead and the number of nozzles, multiple maintenance functions on different parts of the printhead at once It may be desirable to fulfill. To this effect, in such embodiments, it may be advantageous to move the measurement / calibration device around the printhead rather than the reverse. [This then allows other non-optical maintenance processes to be engaged, eg, for another nozzle if desired. To facilitate these operations, the printhead assembly can optionally be “parked” with a system that identifies a specific nozzle or series of nozzles that are subject to optical calibration. Once the printhead assembly or a given printhead is stationary, the motion system 811A is “parked” to precisely position the measurement area 815 at a suitable location to detect droplets ejected from a particular nozzle. “Used to move at least a portion of the optical system relative to the printhead assembly, the use of the z-axis of movement allows selective engagement of the light recovery optics from sufficiently below the face of the printhead. Enable and facilitate other maintenance operations instead of or in addition to optical calibration. Probably stated otherwise, the use of an xyz motion system allows for selective engagement of a droplet measurement system that is independent of other tests or test devices used in the service station environment. This structure is not required for all embodiments and other alternatives where only one printhead assembly is moving and the measurement assembly is stationary or the printhead assembly is not required to be stationed Note that a draft is also possible.

  Generally speaking, the optics used for droplet measurement include a light source 817 and an optional set of light delivery optics 819 (which directs light from the light source 217 to the measurement region 215 as needed); One or more light sensors 821 and a set of recovery optics 823 that direct light used to measure the droplets from the measurement region 815 to one or more light sensors 821. Let's go. The motion system 811A optionally also provides a container (eg, a discharge urn 809) to collect the ejected ink, while measuring droplets from the measurement area 815 around the discharge urn 809 to a subsurface location. Any one or more of these elements are moved along with the discharge crucible 809 in a manner that allows light directing. In one embodiment, the light delivery optics 819 and / or the light recovery optics 823 use a mirror that directs light to / from the measurement region 815 along a vertical dimension parallel to the droplet travel, and the motion system is Each of the elements 817, 819, 821, 823 and the discharge crucible 809 are moved as an integral unit during droplet measurement. This setting presents the advantage that the focus does not need to be recalibrated with respect to the measurement region 815. As described by numeral 811C, the light delivery optics also optionally includes a light source 817 and a light sensor 821 that direct light on both sides of the discharge crucible 809, for example, as generally illustrated. Together, it is used to provide source light from a location below the dimension plane 813 of the measurement region. As described by the numbers 825 and 827, the optical system optionally includes a lens for focusing purposes, and a photodetector (eg, for non-imaging techniques that do not rely on multi-pixel “photo” processing). Can be included. Again, the optional use of z-motion control for the optical assembly and the discharge crucible allows any nozzle at any time during the optional engagement and disengagement of the optical system and the printhead assembly to be “parked”. Note that it allows for precise movement of the measurement area 815 proximate to. Such parking of the printhead assembly 803 and xyz movement of the optical system 801 are not required for all embodiments. For example, in one embodiment, laser interferometry is used to measure droplet characteristics and the printhead assembly (and / or optical system) is deposited on the deposition surface to image droplets from various nozzles. In or parallel to it (eg, in or parallel to face 813). Other combinations and permutations are possible.

  FIG. 8B provides a process flow associated with drop measurement for some embodiments. This process flow is generally specified using the number 831 in FIG. 8B. More particularly, as used by reference numeral 833, in this particular process, the printhead assembly is initially stationed at a service station (not shown) of, for example, a printer or deposition apparatus. The droplet measuring device can then be used, for example, by selective engagement of part or all of the optical system, for example through movement from below the deposition surface to a position where the optical system can measure individual droplets. Engaged with the printhead assembly (835). By numeral 837, this movement of one or more components of the optical system relative to the stationed printhead can optionally be performed in the x, y, and z dimensions.

  As previously suggested, even a single nozzle and the associated nozzle firing drive waveform (ie, the pulse or signal level used to eject the droplet) can vary slightly from drop to drop. , Trajectories, and velocities can be generated. In accordance with the teachings herein, in one embodiment, a droplet measurement system, such as that indicated by numeral 839, can determine that parameter n per droplet so as to derive a statistical confidence regarding the expected nature of the desired parameter. Get a measurement of times. In one implementation, the measured parameter can be volume, while for other implementations, the measured parameter can be flight speed, flight trajectory, nozzle position error (eg, nozzle bending) or another parameter, Or it may be a combination of a plurality of such parameters. In one implementation, “n” may be different for each nozzle, while in another implementation, “n” may be a fixed number of measurements made to each nozzle (eg, “24”) In yet another implementation, “n” is such that additional measurements can be made to dynamically adjust the measured statistical properties of the parameters or refine the confidence level. Refers to the minimum number of measurements. Clearly, many variations are possible. For the example provided by FIG. 8B, assume that the drop volume has been measured to obtain an accurate average representing the expected drop volume and a tight confidence interval from a given nozzle. This average, whether weighted or otherwise, can be assigned by the processor in view of relevant measurement data. This uses multiple nozzles and / or drive waveforms while ensuring the distribution of the composite ink fill in the target area around the expected target (ie, for drop average composite). And) allows optional planning of droplet combinations. As described by optional process boxes 841 and 843, interferometry or shadowography is considered, ideally allowing instantaneous or near-instant measurement and calculation of volume (or other desired parameters). It is an optical measurement process. With such fast measurements, for example, quantity measurements are frequent and dynamic to address changes in ink properties (including viscosity and constituent materials), temperature, power supply fluctuations, and other factors over time. It becomes possible to update to. Based on this point, shadowgraphy typically features image capture of droplets using, for example, a high resolution CMOS or CCD camera as the photosensor mechanism. While droplets can be accurately imaged within a single image capture frame at multiple locations (eg, using a strobe light source), image acquisition is typically (eg, what It involves a finite amount of time so that imaging a sufficient droplet population from a large printhead assembly (in thousands of nozzles) can take several hours. Interferometry, which relies on detection of multiple binary photodetectors and interference pattern spacing based on the output of such detectors, is a non-imaging method (ie, does not require image analysis), so shadowography or It produces drop volume measurements many times faster (eg, 50 times) than other techniques. For example, using a 10,000 nozzle printhead assembly, each large measurement population of thousands of nozzles can be obtained in minutes, ensuring that droplet measurements are made frequently and dynamically. Is expected to. As described above, in one optional embodiment, droplet measurement (or measurement of other parameters such as trajectory and / or velocity) can be performed as a periodic intermittent process, and the droplet measurement system is Engage according to schedule, or between substrates (eg, as substrates are loaded or unloaded), or stacked for other assemblies and / or other printhead maintenance processes. With respect to embodiments that allow alternative nozzle drive waveforms to be used in a manner specific to each nozzle, a high speed measurement system (eg, an interferometer system) is provided for each nozzle and for each alternative. To easily enable statistical population creation for drive waveforms, thereby facilitating planned drop combinations of drops generated by various nozzle-waveform pairs, as previously suggested Please keep in mind. By the numbers 845 and 847, a very precise drop per target deposition area by measuring the expected drop volume per nozzle (and / or per nozzle-waveform pair) to an accuracy better than 0.01 pL Can be planned, complex fillings can also be planned to 0.01 pL resolution, and the target volume is within a specified error (eg, tolerance) range of 0.5% of the target volume or better Can be kept within. As indicated by numeral 847, the measurement population for each nozzle or nozzle-waveform pair is, in one embodiment, to generate a confidence distribution model for each such nozzle or nozzle-waveform pair, ie, , With 3σ confidence (or other statistical measure such as 4σ, 5σ, 6σ, etc.) for acceptable drop intersections. Once sufficient measurements have been made on the various droplets, filling with combinations of these droplets can be evaluated and used to plan printing (848) in the most efficient manner possible. Can do. As indicated by the separation line 849, drop measurement can be performed by intermittently switching back and forth between the active printing process and the measurement and calibration process. In order to minimize manufacturing system downtime, such measurements are typically made while the printer is entrusted with other processes, for example during substrate loading and unloading. Please keep in mind.

  FIG. 8C illustrates another embodiment of a method for droplet measurement, represented generally by the numeral 851. When a printhead is mounted or otherwise desired to calibrate the drop measurement system to compensate for position offsets, to precisely match drop measurements with a given printhead nozzle In the meantime, a calibration routine can be executed. In an exemplary embodiment, the alignment process includes printing from the bottom, i.e., upwards from the perspective of the substrate at the nozzle plate, to identify one or more printhead fiducials or alignment marks (853). This is done using an “upward camera” or other imaging device that captures an image of the head. In one embodiment, the camera (imaging system) can be the same device used for droplet measurement, but can also be a separate imaging device. For example, for an exemplary printhead having 1024 nozzles arranged in 4 rows of 256 nozzles, to determine the offset and rotational tilt between the nozzle plate and the grid system corresponding to the imaging device, The reference on the nozzle plate is used (so as not to be mixed with the substrate reference). In one embodiment, the reference may optionally be a particular nozzle (854), for example, the nozzle closest to the corner of the printhead (eg, first and fourth rows, nozzles 1 and 256), other Note that mechanisms can also be used. In a typical implementation, printhead configuration data (855) is loaded into the system by software and used to identify nozzles at these corners, interpolating (based on first estimating the position of each nozzle). 857) is used to map the address of all nozzles to the grid 856 of the imaging system. By way of example, in one embodiment, the system software is designed to accommodate different printheads with different nozzle configurations, to this effect, the system software can determine the number of rows, if any Load printhead configuration data to identify presence, number of nozzles per row, average vertical and horizontal offset between rows and columns of nozzles, etc. This data allows the system software to estimate the position of each nozzle on the printhead, as described. In one contemplated system, the calibration process is performed once the printhead is replaced, not during the printing process. In different embodiments, the calibration process is performed each time the droplet measurement system is initialized, for example, each new measurement is performed between two printing operations.

  During generation, nozzle (and nozzle waveform) measurements can be made on a rolling basis that advances through a series of nozzles with each interruption during a substrate printing operation. Regardless of whether all nozzles are engaged in a new measurement or such a rolling reference, the same basic process of FIG. 8C can be employed for the measurement. To this effect, the numbers 858 and 859 allow the system software to (for example, move the pointer to the second when the drop measurement device engages for a new measurement (immediately after either the pre-measurement or substrate printing operation). The next nozzle to be measured can be identified (by loading into "Nozzles 2, 312" for the "312th" nozzle of the current print head). For initial measurements (eg, responding to periodic processes such as new printhead installations, recent activations, or daily measurement processes), the pointer is the first nozzle of the printhead, eg, “Nozzle 2,001. Will point to. This nozzle is either associated with a particular imaging grid access or referenced from memory. The system uses the provided address to advance the droplet measurement system (eg, the previously referenced discharge crucible and measurement area) to a position corresponding to the expected nozzle position. Note that in a typical system, the mechanical throw associated with this movement is very precise, i.e. accurate to approximately micron resolution. The system optionally searches the nozzle position for the expected micron resolution position at this point, finds the nozzle, and within a few microns distance from the estimated grid position, that position based on printhead image analysis. (860). For example, a zigzag, spiral, or other search pattern can be used to search for the expected position of the nozzle. (Note that in one embodiment, the process can also be performed manually by adjusting the printer by a human operator.) While the typical pitch distance between the nozzles can be about 250 microns, The nozzle diameter can be about 10-20 microns. Once the nozzle of interest has been identified, the software relies on the droplet measurement system to fire a droplet from the nozzle in question and verify that the nozzle in question has actually fired ( Then, the identification of the nozzle is confirmed). FIG. 8C illustrates the process being performed each time a new nozzle is identified for measurement (eg, each time the droplet measurement system moves), but in some embodiments, during offline configuration Take this measurement once (for example, in a situation where the drop measurement system grid is very tight), store the grid position of each nozzle, and then only when the printhead is replaced or in response to error handling It is also possible to update this position. In droplet measurement systems and / or systems where print head position dynamics are not very precise, it may be advantageous to use an estimate and search function for each nozzle whenever there is a change in the nozzle being monitored. As suggested by numeral 861, in one embodiment, the estimation and search function aligns the drop measurement device (and its associated optics) with the printhead nozzle being monitored in each of three dimensions (xyz). Please note that.

  The precise z position (distance to the drop measurement area) of each nozzle is then adjusted (862) to ensure consistent drop measurement and / or image capture. For example, drop measurement systems typically measure each drop multiple times and calculate these parameters based on distance (eg, relative to the centroid of each drop image) to determine drop velocity and Determining the flight trajectory has been described previously. Various parameters include strobe timing errors (e.g., for shadowography-based drop measurement systems), uncorrected alignment errors between the drop imaging system and the nozzle plate, nozzle process corners, and other factors, Appropriate droplet measurements can be affected. In one embodiment, various statistical processes are used to compensate for such errors, for example, in a manner that normalizes strobe firing for drop measurement locations across all drops. For example, if the hypothetical printhead has 1,000 nozzles, the system will have a desired average in the measurement area (on average over 1,000 nozzles or a subset thereof) with respect to the average drop image position. By choosing an average offset that produces a minimum of position error while centering a few drops, the z-axis offset from the printhead plate can be normalized. Similar techniques can be applied to interferometry-based systems or other droplet measurement systems.

  FIG. 8C shows a droplet measurement region 863 and a hypothetical path for two droplets 864 and 866 through the measurement region along respective hypothetical trajectories 865 and 867. Several should be noted about the example provided by this figure. First, velocity and trajectory measurements are considered to depend on measuring the same droplet multiple times (3 times for droplets 864 and 866, respectively). This requirement may include changing the timing of the strobe (or light source), changing the drive waveform used to fire the associated drop, changing the z-axis position of the drop measurement system, and / or By one or more of changing the z-axis position of the printhead, it can be used to properly position the measurement area relative to the strobe (or imaging source) firing. For example, as the strobe is fired repeatedly (during a single exposure in the case of a shadowography-based system), three droplet images are expected for a single droplet, but only two are observed If so, the measurement area is adjusted to effectively redefine where the drop position is captured relative to the measurement area until the height is misaligned and three exposures are obtained. Inevitably, this hypothesis provides only examples, and other implementations may measure more than three droplet exposures or less than three exposures. Also, the deviation of the trajectory 865 relative to the trajectory 867 may be due to statistical variations in the way droplets are generated, so optionally (in terms of alpha and beta angles) in the average droplet trajectory and each of these dimensions. Note that it can be used to build a statistical model that represents the standard deviation. As should be appreciated, the droplet measurement region 863 shows a two-dimensional droplet depiction (eg, the yz plane, as in the drawing page), while the trajectory angle with respect to the x-axis indicates that the droplet is in FIG. Derived from the change in apparent droplet size within a given image frame during multiple strobe exposures indicating that it is closer to or farther away from the plane of the drawing sheet represented by Can do. Similar interference pattern changes are applied in the case of interferometry-based techniques.

  The scheme represented in measurement 863 can also be used to measure nozzle row bending. That is, as an example, it is assumed that droplets 864 and 866 originate from a common exact nozzle position, but the reverse trajectory is at the expected y-axis center of the droplet measurement region (ie, right to the drawing page). If not aligned (from left to right), the nozzle in question can be offset at its y-axis position relative to other nozzles in the same row or column. As suggested by previous discussions, such anomalies can lead to ideal drop firing deviations that can be taken into account when planning precise combinations of drops, i.e., preferably arbitrary Such row “bends” or individual nozzle offsets are stored and used as part of the print scan plan as previously discussed, and the printing system does not average the differences of each individual nozzle. Use it systematically. In an optional variation, the same technique can be used to determine irregular nozzle spacing along the x-axis, but for such embodiments, any such error is A droplet velocity deviation correction can be included (eg, any such spacing error can be corrected by adjusting the nozzle velocity). In order to determine the y-axis bend of the nozzles that generate droplets 864 and 866, each trajectory 865 and 867 is effectively back-drawn (or otherwise mathematically) with other measurement trajectories of the same nozzle. Applied) and used to identify the average y-axis position of the particular nozzle being monitored. This position may be offset from the expected location of such nozzles, which may be evidence of nozzle row bending.

  As mentioned above, and as suggested by this discussion, one embodiment provides for each nozzle for each parameter being measured, eg, for volume, velocity, trajectory, nozzle bending, and potentially other parameters. Is constructed (868). As part of these statistical processes, individual measurements can be rejected or used to identify errors. In some implementations, if one obtains a droplet measurement having a value that has been removed to some extent from other measurements of the same nozzle, the measurement may represent firing error, in one implementation, the system Discards this measurement if it deviates to a point exceeding the statistical error parameter. If no droplets are seen, this is evidence that the droplet measurement system is in the wrong nozzle (in the wrong position), has a firing waveform error, or the monitored nozzle is inoperable obtain. A measurement error handling process 869 is employed to make appropriate adjustments, including making any new or additional measurements as needed. By the number 870, each measurement is advantageously stored and used to build an associated statistical distribution, and the system then uses the same nozzle until sufficient robustness to measurement error is obtained. Loop to make additional drop measurements from. This loop (871) is considered in FIG. 8C to show the implementation until n droplets are obtained for each nozzle or each nozzle-waveform pair. When a sufficiently robust distribution is obtained, the system calculates (stores) the desired statistical parameters (eg, average, standard distribution of each scale parameter), assigns it to a given nozzle (872), and arbitrarily Perform an appropriate error handling process 873 (just verify the measured nozzle or consider it or the associated waveform inoperative) and then proceed to the next firing waveform or next nozzle as appropriate (874). . That is, once the measurement distribution for a given nozzle or nozzle-waveform pair is complete, the system software identifies the next nozzle to be measured (874, eg, updated by the address pointer), and then by the number 876 Return to move the drop measurement system as appropriate, or to take the next measurement. Alternatively, if the number 875 indicates that the time has come and the system is required to print another board as part of the production line, the system will use the newly acquired data (if any). The arbitrary scan operation is updated based on the above, the address of the “next” nozzle is stored, and the process returns to the substrate printing (875). In one embodiment, after such a printing operation is completed, during an expected interruption (or maintenance downtime), the system reads the stored nozzle address and droplet measurement details, which are interrupted. Continue from the beginning.

  Note that although not separately invoked by FIG. 8C, the depicted measurement process will typically be performed on each alternative waveform available for use with each nozzle. For example, if each nozzle had four different piezoelectric drive waveforms that could be selected, the inner process loop 871 of FIG. When seeking to build a statistical distribution based on drops, there can be 96 such measurements per nozzle (24 for each of the four waveforms), each measurement being the drop velocity, trajectory, And volume respectively, and (for example, for purposes of assessing nozzle bending) are used to generate a statistical average and a diffusion measure of estimated nozzle positions.

  FIG. 8D shows a flow diagram for nozzle qualification. In one embodiment, a statistical model (eg, distribution and average) for each and / or each of drop volume, velocity, and trajectory for each nozzle and for each waveform applied to any given nozzle. Droplet measurements are taken to produce Thus, for example, if there are two selections of waveforms for each of the 12 nozzles, there are up to 24 waveform / nozzle combinations or pairs. In one embodiment, each parameter (eg, volume) measurement is made for each nozzle or waveform-nozzle pair sufficient to generate a robust statistical model. Despite planning, it is conceptually possible that a given nozzle or nozzle-waveform pairing can yield an exceptionally wide distribution, or an average that should be specially treated because it is sufficiently abnormal Please note that. Such special treatment applied is conceptually represented in FIG. 8D in one embodiment.

  More particularly, reference numeral 881 is used to represent the general method. Data generated by the droplet measurement device is stored in memory 885 for later use. During application of method 881, this data is retrieved from memory and the data for each nozzle or nozzle-waveform pair is extracted and processed individually (883). In one embodiment, each variable is considered eligible such that a normal random distribution is represented by the mean, standard deviation, and number of droplets measured (n), or using an equivalent measure. Built for. Again, note that other distribution formats (eg, Student's T, Poisson, etc.) may be used. The measured parameter is compared to one or more ranges (887) to determine whether the relevant droplet can be used in practice. In one embodiment, at least one range is applied to consider the droplet ineligible from use (e.g., if the droplet has a volume that is sufficiently large or small for the desired target) Nozzles or nozzle-waveform pairs can be excluded from short-term use). To provide an example, if a 10.00 pL droplet is desired, for example, a droplet that is more than 1.5% away from the target (eg, <9.85 pL or> 10.15 pL) Nozzles or nozzle waveforms that are tied to the average can be excluded from use. Alternatively or alternatively, a range, standard deviation, variance, or another diffusion measure can be used. For example, if it is desired to have a statistical model of droplets with a narrow distribution (eg, 3σ <1.005% of the mean), droplets with measurements that do not meet this criterion are excluded. Can. It is also possible to use a sophisticated / complex set of criteria that takes into account multiple factors. For example, an abnormal average combined with a very narrow diffusion may be accepted, eg, if the diffusion (eg, 3σ) away from the measured (eg, abnormal) average μ is within 1.005% The relevant droplet can be used. For example, if it is desired to use a droplet with a 3σ amount within 10.00 pL ± 0.1 pL, nozzle-waveform pairing produces a 9.96 pL average with a 3σ value of ± 0.8 pL Can be excluded, but nozzle-waveform pairs that produce a 9.93 pL average with a 3σ value of ± 0.3 pL may be acceptable. Clearly, many possibilities are possible according to any desired rejection / abnormality criteria (889). The same type of processing can be applied to the flight angle and velocity per droplet, i.e. the measurement of the flight angle and velocity per nozzle-waveform pair exhibiting a statistical distribution and derived from a droplet measurement device. Note that depending on the statistical model, it is expected that some drops may be excluded. For example, droplets having an average velocity or flight trajectory that is outside of the normal 5%, or a velocity distribution that is outside a specific target can be hypothetically excluded from use. Different ranges and / or metrics can be applied to each droplet parameter measured and provided by the storage device 885.

  Note that depending on the rejection / abnormality criteria 889, the droplets (and nozzle / waveform combinations) may be treated and / or treated differently. For example, as described, certain droplets that do not meet the desired norm can be rejected (891). Alternatively, additional measurements can be selectively performed for the next measurement iteration for a particular nozzle / waveform pair, and as an example, if the statistical distribution is too wide, through additional measurements Special measurements can be made for specific nozzle / waveform pairs to improve the tightness of the statistical distribution (eg, variance and standard deviation depend on the number of data points measured) To do). The number 893, for example, to use a higher or lower voltage level (eg, to provide a higher or lower speed, or a more consistent flight angle), or a tuned nozzle waveform that meets specific criteria It is also possible to adjust the nozzle drive waveform to shape the waveform to produce a pair. The number 894 can also adjust the timing of the waveform (eg, to compensate for the abnormal average velocity associated with a particular nozzle / waveform pairing). As an example (previously suggested), a slow drop can be fired at an early time relative to other nozzles, and a fast drop at a later time so as to compensate for a faster flight time. Can be fired. Many such alternatives are possible. Finally, the number 895 allows any adjusted parameter (eg, firing time, waveform voltage level or shape) to be stored, optionally one adjusted parameter if desired. Or it can be applied to re-measure related drops above it. After each nozzle-waveform pair (modified or otherwise) is deemed eligible (passed or rejected), the method then proceeds to the next nozzle-waveform pair by the numeral 897. Proceed to Once again, specific droplet details are advantageously considered in deriving the print grid firing instructions to ensure the homogeneity of any transformed deposition parameters (at least on a local basis). . In one embodiment, nozzle details are weighted into the transformation calculation for each grid point. In another embodiment, a printed grid spot firing is used with a second process used to extract, redistribute, or otherwise adjust firing decisions made for grid points corresponding to abnormal droplets or nozzles. The decision is made on a weighted criterion that depends on the duplicate template print image overlap (as discussed above). In other words, the nozzle firing decision can be made in the first conversion process, and then the second error correction process can be applied to take into account nozzle or droplet details. Other alternatives are possible.

  Through the use of precision mechanical systems and droplet measurement system alignment techniques, the disclosed methods are described parameters (eg, volume, velocity, trajectory, nozzle position, droplet deposition position, nozzle bending, and other parameters). Enables very precise measurement of individual nozzle characteristics, including the respective average droplet metrics. As should be appreciated, the described techniques facilitate a high degree of uniformity in the manufacturing process, particularly the OLED device manufacturing process, and thus increased reliability. In particular, the teachings presented above are flexible in the fabrication process by providing control efficiency with respect to the speed of droplet measurements, the stacking of such measurements on other system processes, and the incorporation of alignment error correction processes. It helps to provide a faster and cheaper manufacturing process that is designed to provide both accuracy and precision.

  FIGS. 9A-9C are used to illustrate an exemplary runtime printing process for a product array, again giving examples of one or more flat panels on a substrate.

  FIG. 9A depicts a substrate 901 with a number of dashed boxes representing individual panel products. One such product, found in the lower left of the figure, is designated using reference numeral 902. Each substrate (in a series of substrates) has a number of alignment marks, as represented by numeral 903, in one embodiment. In one embodiment, two such marks 903 are used on the substrate as a whole, allowing adjustment of substrate position offset, rotation error, and scale error, and in another embodiment, three or more of that Such a mark 903 is used to facilitate adjustment of the tilt error. In yet another embodiment, each panel (such as any of the four depicted panels) is accompanied by a per-panel alignment mark, such as mark 905. These alignment marks can be used to perform independent error handling per panel or per product. Once again, a sufficient number or density of such marks (eg, 3 or more per panel or other product) allows for compensation of non-linear errors. There may be these marks in addition to or instead of the substrate reference 903. In yet another embodiment, the alignment marks are replicated at regular intervals regardless of panel position (eg, as represented by oval 909). Whichever scheme is used, one or more cameras 906 are used to image the alignment marks to detect the errors referenced above. In one contemplated embodiment, a single stationary camera is used so that the printer transport mechanism (eg, handler and / or air floating mechanism) positions each alignment mark in turn within the field of view of the single camera. Move the substrate. In different embodiments, the camera is mounted on the motion system for transport relative to the substrate. In yet another embodiment, as discussed below, a low or high magnification image that identifies coarse reference positions according to a printer coordinate system and a low magnification image that coarsely positions a reference for high resolution magnification A high-magnification image is taken. In light of the previous discussion, in one embodiment, the printer transport mechanism controls movement to within about 1 micron of the intended position, so the system can accurately track the position in software. The error can be calculated for a printer-based or substrate-based coordinate system.

  In a typical implementation, printing is performed to deposit a given layer of material over the entire substrate at once (ie, using a single printing process that provides layers for multiple products). Let's go. To illustrate this, FIG. 9A shows two illustrative scans 907 and 908 of the printhead along the long axis of the substrate, and in a split axis printer, the substrate is typically printed during the scan. It is moved back and forth (eg, in the direction of the depicted arrow) with a printer that advances the head positionally (ie, perpendicular to the drawing page). Note that although the scan path is depicted as linear, this is not required in any embodiment. Also, although the scan paths (eg, 907 and 908) are illustrated as adjacent and mutually exclusive with respect to the coverage area, this is also not required in any embodiment (eg, the printhead can print as needed). Can be applied on a partial basis to the band). Note that, ultimately, any given scan path typically passes through the entire printable length of the substrate, so as to print layers for multiple products in a single pass. . Each pass uses a nozzle firing decision according to the printed image (as distorted or corrected), and each firing decision has a selected and programmed waveform to produce the desired droplet volume, trajectory, and velocity. Apply. Advantageously, the processor built into the printer (as previously introduced) allows both nozzle / droplet measurement and parameter qualification and update, detection of details per substrate or per panel, template Nozzle firing data (eg, “data” and “from FIG. 5B” that will be used in connection with corrections and binary nozzle firing decisions (see, eg, the “trigger” signal from FIG. 5B). (See “Drive Waveform ID”). Once printing is complete, the substrate and wet ink (ie, deposition liquid) can then be transported for curing or processing of the deposition liquid into the permanent layer. For example, briefly returning to the discussion of FIG. 6B, the substrate breaks the all controlled atmosphere (ie, advantageously used to prevent moisture, oxygen, or particulate contamination). Without having the “ink” applied in the printing module 625, it can then be transported to the curing chamber 641.

  FIG. 9B again illustrates one alignment and detection process 911 for manufacturing operations, using an example of flat panel device fabrication. Note that many alternative processes are possible, and FIG. 9B provides an example only. The process is a three-part operation method in which a new substrate is first roughly mechanically aligned, then the alignment mark is optically measured, and finally the software process corrects the fine (virtual) alignment. Is used. When a new substrate is loaded, the robot first places the substrate on printer lift pins that are used to advance the substrate to the vacuum gripper. The gripper moves the substrate while being supported by the air floating table, and the substrate is roughly aligned and in place for optical measurements (ie, to identify a reference location) Advances to position. The system is programmed for coarse substrate details so that a camera control system (or other imaging device) is activated and / or activated as needed to image areas where the presence of a reference is expected. Note that this is done. First, a low magnification image is captured, the substrate is repositioned, and then a higher magnification image is used to accurately detect the reference position. During either phase of the optical measurement, a search process (eg, a helix) is used to find each criterion as appropriate. The system then continues with a second criterion (and / or an appropriately higher order criterion, for example) to help identify the precise position, orientation, tilt, and / or scale of the substrate, as appropriate. Measure. As described above, the process can also be applied on a panel-by-panel basis or in fact to any and all standards to obtain position / distortion information at the appropriate resolution for the associated error correction process. The detected reference parameters (such as position, size, shape, rotation, tilt, or related reference reference points) can be compared with the expected parameters of such a reference to detect errors. As described, any detected error may involve one or more of translation, rotation, scaling, or tilt errors, or a superposition of a plurality of such errors. Based on the detected error, printer control data is constructed that corrects any error so that the print is in the “right place”. As described on the right side of the figure, based on the alignment mark determination, the system software corrects the error by an action on the template. As described above, the template is read out, and then an instance of the template is appropriately manipulated to generate an appropriate nozzle firing decision via the error correction process described above (or another equivalent process) ( Rendered). Thereafter, printing is performed.

  FIG. 9C provides an illustrative schematic that is used to discuss substrate error correction in software. In one embodiment, generally represented by numeral 921, data representing a layer to be printed is first read (923), eg, according to a substrate or panel recipe, or both. As described by the numeral 925, this information is typically maintained as a template, a copy or “instance” of the data is read, fitted to the error, sent to the printer, and then (ie, Discarded during next substrate process (to avoid data contamination from previous substrate processing). These processes are variously represented in FIG. 9C.

  With a copy of the template at hand, the system detects the substrate geometry by the numeral 927. As described by the numbers 928-932, the detection process may be interrupted once (eg, before printing begins or when a new substrate is loaded) (eg, the printing process may be interrupted). Or reference capture can be performed for each of a plurality of subdivisions of the substrate, for example, for each panel), can be repeated, or can be performed on a continuous basis. As described in detail by numeral 931, in one embodiment, a plurality of different criteria are used to allow detection and correction of different types of linear and non-linear errors. Once an error is detected, then by the number 935, the system calculates a transformation. For example, in one embodiment, because the error is linear across the board, a simple linear equation is derived and used to render the cached template to generate printer control data. In other embodiments, the error can be modeled by a quadratic equation or other polynomials in a manner that is discontinuous (eg, according to a region) or in some other manner. In a parallel processing embodiment, a master or supervisory processor makes this determination and then assigns processing to a discrete core or processor. The numeral 937 then causes the system to subsequently find relevant data from the read instance of the template that it uses to correct and / or assign nozzle firing decisions to mitigate errors. In the embodiment where the template is in the form of a bitmap of nozzle firing data, as previously discussed by the numbers 938-941, the system will fit the error to the pixel “closest” to the error vector position in the template. A weighted measure of multiple pixels from the template or in some other manner can be the basis. The numeral 940 allows an affine transformation to be applied using a process that essentially performs a matrix operation on the “tile” of printed grid points to obtain a new launch decision. For example, such transformations may weight template data by offset, rotation, and other factors to obtain “transformed space” data corresponding to true substrate position (and orientation, tilt, etc.) Can do. Other processes can also be performed with the number 941. Once nozzle assignment has been made, the system can then optionally also invoke post-position compensation processing to correct for discrete area filling or ink density errors on the substrate. Several options and types of processing are represented in FIG. 9C (which will also generally be discussed below in connection with FIGS. 10A-E). For example, the numeral 951 illustrates a hypothetical fluid well that can hold a light generating element of an OLED display panel with a particular node or droplet position indicated by the numeral 954, overlaid by a printed grid 953. If the position error results in too many or too few drops within the boundaries of the well 951, an anti-aliasing process can be applied to check the number of drops that will fit in the well. . For example, rotation of the printing grid can change the number of droplets that will fit in the well given the droplet density represented by the template, and the anti-aliasing process will detect this problem, The nozzle assignment is modified (945) from the position compensation process so that the correct number and / or volume of droplets is deposited in the well. Droplet details and / or nozzle verification can also be taken into account for the process, as shown on the right side of the figure via the numbers 946-948. For example, a droplet at node 954 stores a droplet parameter that indicates a position error (eg, the error that an expected firing trajectory would accrete outside the fluid well delineated by the droplet) the problem also considers the details of other “nodes” that fit within the well to determine which should have changed the firing decision to meet the required filling parameters It can also be corrected by software (945). Similarly, if a validation process is used and a firing decision is assigned to an inoperable (or deemed ineligible) nozzle by the position compensation process, the problem can be detected and corrected by the software (945). ). As mentioned above, in yet another embodiment, droplet parameters are measured in situ and periodically updated to provide a robust current data set that takes into account changing conditions. it can. Generally speaking, step 945 corrects for emphasis and aliasing errors to ensure that the liquid delivery will meet the necessary details for deposition. If there are errors detected by process 955 that cannot be handled or taken into account, exception handling routines (957) typically deal with the error in question or refresh the data. Called to adjust or recalculate the scan path to render. Then, if the error is completely mitigated, the output data is stored as appropriate and / or transmitted to the printer and printhead by the numeral 959. In one embodiment, the system software further detects a repetitive error that correlates between the substrates, and then optionally “learns” the repeatable error and adapts the template to that error (ie, Note that the cached template can be updated (by 961). For example, an unintended protrusion of the system edge guide for the substrate that slightly “fluctuates” the substrate at or with respect to a particular position in its motion can cause position errors in a given system. It can be caused. These and other repeatable errors can be learned by the system and used to reduce job-by-job error handling. This technique will be further discussed below in connection with FIG. 10E.

  FIGS. 10A-10E show different types of errors that can be applied differently, ie, in assigning nozzle firing decisions, in adjusting droplet details for a particular node of a printing grid, and / or as a post-position compensation process. Used to indicate compensation technique.

  More specifically, FIG. 10A shows a flowchart 1001 for droplet allocation and / or droplet adjustment in a region. For example, such techniques can be applied to a halftoning on a local basis (ie, to deposit a fluid that can diffuse to provide a blanket coating with a calibrated thickness), or (eg, to FIG. 9C Can be applied to the total filling in the fluid wells (as discussed above in relation). The method first identifies the region or well in question by the numeral 1003 and identifies the desired number of drops or the desired fill volume. The number 1005 then causes the system to read the statistical mean and variance of the droplet parameters for each nozzle and / or nozzle drive waveform. This data provides an understanding of the expected droplet parameters with some confidence. As represented by the parameters v, α, and β, in one embodiment, the system includes from memory a parameter that represents an expected droplet volume and an expected two-dimensional droplet accretion location, and these parameters. Note that the variance of The system software then continues to mimic filling to determine whether the expected ink density (or total volume) meets a predetermined threshold. In one embodiment, the system simply determines an eligible combination of droplets needed to meet a predetermined threshold and then (ie, a plurality within a scan band traversed by the printhead) Select one of these eligible combinations (in a manner that optimizes the ability to do this simultaneously for the region). In another embodiment, the system simply selects a number of droplets and then selects the combination if the combination of droplets would be expected to produce a result outside a predetermined threshold. Adjust after the fact. By the numbers 1006, 1007, and 1008, the system, in one embodiment, may have different drive waveforms for a given nozzle (eg, one of 16 pre-programmed waveforms as previously described). ) Can be selected, or a different nozzle can be selected, or a check can be made to see if a particular nozzle is available. With the number 1009, the system adjusts the nozzle assignment (including drive waveform selection information) as necessary, and then the printing process is again scrutinized for errors (1011, eg, predetermined scan path and scan). Of number). If there is an error, the system can adjust the printhead offset for a given scan path (1012) and / or adjust the number of scans (1013). If there is no error, the method ends (1014) and the assigned data is output to the printer as previously discussed.

  FIG. 10B provides another flowchart 1021 regarding template adjustment, this time as a step after error correction. By the number 1023, for N regions or fluid wells, the system software subsequently identifies the well location, for example, based on detected error data. In step 1025, the software reads the firing decision of the printed grid nodes that fall within the region of interest and reads the mean (μ) and standard deviation or other diffusion measure (eg, σ) of each such parameter (1026). Following the simulation (1027) of the total volume or volume distribution of the region, by the numbers 1029 and 1031, the system then determines that the volume or distribution has a predetermined threshold, eg, a minimum threshold (Th1) and a maximum threshold. It is determined whether it matches (Th2). If the volume or distribution does not match the predetermined threshold, the system software continues to drop and / or as indicated by the numbers 1033, 1035, and 1036 until the volume or distribution matches the predetermined threshold. Adjust waveform assignment. In one embodiment, if multiple waveforms are available for each nozzle and are pre-selected to produce intentional variations in droplet volume, simply assign a new firing waveform to a particular printed grid node. Thus, the error can be compensated without changing the scan path. If this waveform is different from another firing decision for the same printhead nozzle, a different “default” waveform for a particular nozzle during nozzle firing (ie, “trigger (firing) depicted in FIG. 5B It is up to the system to schedule the selection) The number 1037 then causes the system to save relevant nozzle data as appropriate. If there is no suitable waveform programmed for a particular nozzle, as described by optional process block 1039, in one embodiment, the system will attempt to improve the firing details obtainable from a given nozzle. At this time, the waveform can be added or the selected waveform can be reprogrammed.

  FIG. 10C shows a process flowchart 1041 in which nozzle / waveform data is taken into account for rendering an instance of a template to account for substrate or panel errors. More particularly, as indicated by the numbers 1043-1047, the system can read nozzle drive details corresponding to a bitmap (eg, a template in the form of a bitmap). For example, again, using the example where the new printed grid node firing assignment is based on a weighted average of the four printed pixels corresponding to the template, the four assigned nozzles are 10.000, 10.50, respectively. If expected to produce 10.00 and 10.20 pL droplets, this information can be used to select either nozzles or nozzle waveforms to deposit the droplets on misaligned substrates. Can be used, considered. Returning to the fluid well 951 embodiment from FIG. 9C, any printed grid node (eg, 954) that fits within the depicted well and is assigned to fire is potentially a total for the fluid well. The system can be used to obtain a volume, and the system effectively weights four printed pixels in this hypothesis and has a volume parameter with the expected volume corresponding to the weighted combination of four printed pixels. New grid points (eg, nozzle or firing time) and associated drive waveforms (if multiple waveforms are available) can be selected. The numeral 1049 causes the software to apply any smoothing (eg, adjustment of adjacent print pixels or next drop selection as needed) as desired, and again output data (selected drive waveform). Can be written to memory. This output data will then eventually be sent to the printer to control printing. The number 1053 allows the system to assume a fixed scan pattern (ie, simply assign nozzles and drive waveforms in a manner that addresses errors), or in another embodiment, all minimize print time. Looking at the scan, the scan (rasterization) can be rethought and optimized again, for example, by changing the scan path, printhead offset, number of scans, or other details. These optional features are represented by the numbers 1055 and 1057.

  FIG. 10D shows yet another flowchart 1061 for halftone (ie, droplet density) adjustment. In connection with this embodiment, for example, to provide a blanket coating over the area of the substrate using the deposited ink, but the ink droplets are of a desired layer thickness (ie, a predetermined limit for each droplet). It would be assumed that it would be desirable to control the ink drop density so that it is deposited at a density that imparts a given diffusion characteristic). Such a process is particularly useful for creating barrier layers, encapsulation layers, or other layers where the layer thickness is desired to be consistent over a relatively large area. The numeral 1063 once again assumes that some portion of the data from the template is to be rendered to compensate for substrate and / or panel position errors. By the numbers 1067, 1069, and 1071, the software is required in this embodiment to produce the expected droplet volume, position and dispersion, and the desired layer parameters (eg, thickness, and desired thickness). Read halftone density or ink density). The system then calculates the nozzle and drive waveforms (eg, a specific scan path assuming a specific scan path) from the available pool in a manner calculated to facilitate the same ink density represented by the original template data. Select a nozzle that will pass through the area. For example, if it is desired to produce a 5 micron thick encapsulation layer, the system will: (a) a nozzle that will pass through each region, the desired required to produce the desired thickness The ink density, per-nozzle and per-waveform droplet details; Apply mathematical functions to select, then output printer control data. When planning a halftone or density pattern, the system software typically plans the deposition carefully to deposit a fluid of uniform density, eg, consistent volume (considering nozzle details) For example, to balance a “light” droplet (eg, 9.00 pL) and a “heavy” droplet (eg, 11.00 pL) in terms of droplet distribution when 10.00 pL) is not available Please keep in mind. Once again, the number 1073 causes the output details to be scrutinized again for errors (eg, the scan path cannot produce the desired result) and if no errors are found, the data is output to the printer (1075) or vice versa. If an error is found, an exception process is invoked (1077, eg, to reassess rasterization). Other alternatives will occur to those skilled in the art.

  Finally, FIG. 10E provides a flowchart 1081 relating to updating a stored template for repeatable errors, for example, as introduced above in connection with FIG. 9C. For each new substrate, the depicted method compares the deviation from the expected position (1083) with the past error stored in memory by the numeral 1085. The system software attempts to detect a correlation in error (e.g., as opposed to a unique substrate error) by the number 1087. By the numbers 1088-1090, the degree of correlation that is determined may be based on some type of calibration (ie, to set the expected board norm), continuous monitoring of continuous board criteria, or inferred calculations. The system uses numbers 1091 and 1093 to effectively “learn” patterns in error, for example, based on regression software, neural nets, or other adaptive processes, and then modify the template accordingly and use deviation data. Update the history of. Finally, with the number 1095, the software corrects the current board for any uncorrected error (eg, unique error) and outputs the data to the printer as previously described.

  As mentioned above, it is generally desirable to print quickly to minimize processing time and increase throughput. In one embodiment, a substrate that is approximately 2 meters in width and length can have a layer uniformly deposited over its surface in less than 90 seconds per layer. In another embodiment, the time is 45 seconds or less. Thus, in such applications, it is advantageous that rendering and per-board (or per-product) error reduction occur as quickly as possible.

  11A-11B are used to introduce a parallel processing architecture that facilitates this processing speed. A detailed print grid that generates printer control data using precision manufacturing (for example, printing many television screen layers, each with millions of pixels), as should be apparent The software adjustment can take several seconds. The architecture and process presented in FIGS. 11A-11B are ideally used to reduce this time to 2 seconds or less.

  More specifically, as represented in FIG. 11A, a copy of the cached print image, stored recipe data, or other relevant source data representing ideal printing is first read. Since the master or supervisory processor 1103 will already be aware of any board configuration data, it can understand the number of products (panels) and their respective expected locations. A monitoring processor 1103 receives the alignment data and calculates an error (eg, for each panel or product on an independent basis). Panel processing, or error compensation at the sub-panel level, is optional, i.e., the disclosed technique is applied even when instances of the printed image are matched on a basis that is linear throughout a particular substrate. Can be done. The system also includes a set of parallel processors 1105, in one embodiment, a multi-core processor or graphics processing unit (GPU). A GPU typically includes a large number (eg, hundreds to thousands) of cores 1107, each (or many of them) utilized in this embodiment for processing in parallel. The supervisory processor 1103 determines how many parallel threads or processes should be used to convert the template print image instance, depending on the panel definition, desired error correction, and other factors. Make a decision. In connection with the process, the supervisory processor subdivides the overall print area of the board as appropriate, assigns any relevant conversion parameters to the relevant core or processor, and stores (caches) an instance of the template print image in the embedded DRAM 1109. ). Embedded DRAM is advantageously used by one or more of cores 1107 for error correction and rendering of printer control data (including conversion and overlay of any source data), Note that it includes a plurality of banks, ports or arrays 1111 each having a register 1111A. Embedded memory is advantageously structured for parallel access so that access by one core or processor does not limit the bandwidth of another core or processor. To this effect, the architecture features a very wide access path 1113 (eg, featuring hundreds of data access lines as appropriate) so that each core can access memory 1109 in parallel. be able to. In one implementation, the manner in which source data is stored is determined by the processor or core allocation so that each core or processor can access the data needed for conversion and the write area for converted printer control data. Note that it depends on. In one contemplated implementation, the monitoring processor 1103 does not necessarily use all available cores or processors, but does not use cores or processors that handle discrete products (eg, panels), or mutually exclusive portions of panels or products. assign. For example, if a hypothetical panel has 15 panels and the system includes 31 parallel processing units, the supervisory processor may assign two parallel processing units, each unit processing half of the panel. The supervisory processor directs the portion of the template to the embedded DRAM for appropriate storage, and any conversion parameters (eg, gradient algorithm) for each memory bank or array (ie, for each core or parallel processor). ) Write to the relevant register 1111A and then wait for a signal from each core or processor that the process is complete. The output is then transformed into printer control data that is distorted to adapt the printing to any substrate or product variation to precisely position the deposited layer of the material in question. A set for each substrate. As represented by the numbers 1115 and 1117, each of the supervisory processors and their respective cores or processors has their respective supervisory processors, cores, or processors configured to perform specific processing functions as previously described. Note that it is governed by the instructions stored on the controlling, non-transitory machine-readable medium. Also, the rendered data is written to memory by one processor core (as it is available), and other data is obtained by independent means (ie, one processor core follows to correct the position error). Note that embedded DRAM can be designed to provide a form of direct memory access (DMA) so that it can be unloaded (while adjusting).

  FIG. 11A shows some additional implementation options on the right side of the diagram. First, as described above, each panel can be assigned to one or more respective cores or processors, as represented by numeral 1119. As in a typical manufacturing process, a series of substrates use the same recipe or template, and the configuration represented by option 1119 is typically a fixed cost (eg, the monitoring processor 1103 typically (There is no need to recalculate assignments or change the data storage parameters of each board). Second, the number 1120 allows the supervisory processor to optionally assign each transformation per core, taking into account this fixed assignment. While the above examples generally discuss separately offset, rotation, scale, and tilt corrections, in a typical implementation, the transformation may be complex with any combination of these corrections as appropriate. The supervisory processor computes the relevant transformation and then assigns the computed transformation to the appropriate processor or core that performs the transformation on all affected template data to which it is assigned. As represented by numeral 1120, in one embodiment, each transformation is provided to a respective processor or core. The number 1123 allows the storage of preprocessed template data or transformed data to be distributed in one embodiment. That is, it has been previously described that the transformation may involve weighting the firing decision of duplicate pixels on the (transformed) print image or print image data representing the panel. With the option represented by the numeral 1123, the print image pixel data can be processed by storing the northwest, northeast, southwest, and southeast pixel data in the same memory row in their respective memory arrays (rather than at the same address), etc. Can be stored in a manner that contributes to or in another manner. There are many possible memory storage techniques that can be used to accelerate data retrieval and processing (eg, as is conventionally done for graphics and gaming processes, to accelerate image rendering). . The same type of memory process can be employed in this embodiment. Also, as represented by numeral 1123, in one embodiment, for example, if print pixel data is stored in a given row, the system automatically sets the next set of data required for processing. A variable column offset (per memory array) can be used so that it can be fetched quickly. As an example, the cached print image data is distributed in a distributed manner (eg, each of the two memory arrays, with column offsets used to read the associated print pixel information for both memory arrows on an incremental basis. Middle row) can be stored in DRAM (1111). Many alternatives and variations of this process are possible.

  Note that the configuration described above is not the only possible. For example, instead of assigning a board terrain to each core, the supervisory processor or other master 1103 may split the transformation operation and assign a different process to each core (1121). In one embodiment, one core may be assigned to perform one task associated with an affine transformation, while a different core may be assigned to perform another task. As indicated by numeral 1122, almost any assignment of responsibility can be affected among multiple cores, whether parallel or sequential. Generally, it is desirable to proceed with printing as quickly as possible to detect per-board or per-panel errors and maximize manufacturing throughput, so if any efficiency of acceleration meets this objective, It can potentially be applied to multiple available processors or cores.

  FIG. 11B shows a flowchart 1151 associated with the runtime process. In this example, once the error is measured, each of the multiple cores is assigned to apply one or more affine transformations to each terrain of the overall printed area (ie, of the substrate). It will be assumed. As indicated by numeral 1153, template data is first loaded or cached. The surveillance processor utilizes such source data as appropriate (1155) to identify (1157) the relevant alignment mark or reference, and then directs the printer camera system to the appropriate coarse coordinates. The printer camera system returns precision measurement information that the monitoring processor measures against the source information 1155 obtained. The monitoring processor then performs panel and / or board conversion calculations as appropriate, taking into account the detected terrain (1159), and then shares the processing range and appropriately assigns parameters to each core, processor, or parallel process. Convert to threads (1161 and 1163). For repetitive substrate recipes, the monitoring processor may automatically analyze the received subset of template images into memory dedicated to a particular core, processor, or thread. Each core, processor, or thread then begins transforming its assigned data (1165) to transform or distort the template in a manner that matches the print details to the detected substrate or panel terrain. As part of this process, considering any printhead, droplet, or nozzle details, to ensure that the finished converted printer control data can provide accurate deposition, Latest print nozzle and / or drop data is made available (1167). In other embodiments, this adjustment (ie, depending on the print nozzle details) may be performed by a dedicated core, processor, or thread (ie, separate from the core, processor, or thread that performs the print grid operation). Note that this can be done by the monitoring processor. Once any conversion has been performed, the converted printer control data is unloaded from memory, rasterized (1169) and appropriate for printing (1171), consistent with nozzle / droplet specific characteristics. Can be sent to the printer using a simple command.

  Considering the various techniques and considerations introduced above, a manufacturing process can be performed to rapidly mass produce a product at a low per unit cost. When applied to display device manufacturing, eg, flat panel displays, these techniques optionally allow for a high-speed panel-by-panel printing process where multiple panels are generated from a common substrate. By providing a high-speed repeatable printing technique (eg, using a common ink and printhead for each panel), printing can be substantially improved, for example, fill deposition per target area is all within specification It is believed that the printing time per layer is reduced to a fraction of the time that would be required without using the above technique. Returning again to the example of a large HD television display, each color component layer represents a substantial process improvement for 180 seconds or less, or even 90 seconds or less, for a large substrate (eg, about 220 cm × It is believed that it can be printed accurately and reliably for an 8.5 generation substrate that is 250 cm). Improving printing efficiency and quality opens the way for a significant reduction in the cost of producing large HD TV displays and thus lower end consumer costs. As mentioned above, display manufacturing (and in particular OLED manufacturing) is one application of the techniques introduced herein, but these techniques can be used in a wide variety of processes, computers, printers, software, manufacturing. It can be applied to equipment and end devices and is not limited to display panels. In particular, the disclosed technique allows a printer to deposit multiple product layers as part of a common printing operation, including but not limited to any microelectronic, microoptic, or “3D printing” application. It is anticipated that it can be applied to any process used to

  Note that the described technique provides a number of options. In one embodiment, panel (or product-by-product) misalignment or distortion can be adjusted on a product-by-product basis within a single array or on a single substrate. Despite the common print data, the scan path across the two panels has a different firing command for each substrate (eg, rotation or adjustment of data on one panel can vary from print job to print job). Thus, the printer scan path can be planned with the next adjustment / adaptation based on one or more alignment errors. Optionally, this information can be adjusted from a source template (eg, a bitmap representing a binary launch decision) in real time. In other embodiments, print areas and / or scan paths can be added or completely replanned per substrate, despite the common printer source data. The described techniques can be used to make OLED panels, for example 2, 4, 6 or different numbers of panels, as part of a single print job. Following fabrication, these panels can be separated and applied to each product, for example, to produce a respective HDTV display or other type of device. By performing fine alignment (e.g., sub-millimeter alignment) in software, the disclosed technique enables precise mechanical positioning and deposition errors in a manner consistent with underlying product or substrate misalignment or deformation. The precise placement and alignment of the product is not so important, and more accurate product production is provided.

  In the foregoing description and accompanying drawings, certain terminology and drawing symbols are set forth to provide a thorough understanding of the disclosed embodiments. In some instances, the terms and symbols may suggest specific details that are not required to practice these embodiments. The terms “exemplary” and “embodiment” are used to represent an example rather than a preference or requirement.

  As shown, various modifications and changes may be made to the embodiments presented herein without departing from the broader spirit and scope of the disclosure. For example, any feature or aspect of an embodiment may be applied in combination with any other of the embodiments or in place of its corresponding feature or aspect, at least where practical. Thus, for example, not all features are shown in every and every drawing, for example, features or techniques shown in accordance with an embodiment of one drawing are optionally declared in detail herein. It is also assumed that it can be employed as an element of or in combination with features of any other drawing or embodiment. The specification and drawings are, accordingly, to be regarded in an illustrative sense rather than a restrictive sense.

Claims (1)

Invention described in this specification.